February 14, 2016 / Dennis Holinka
Topic 3 – Information Architecture layer
This week's posts go over the Information Architecture layer, the various perspectives, and related reflections in the blog.
Post 2 - Enterprise Information Ontology – the foundation for EIA, EMM, EIM
The work of creating an Enterprise Information Architecture (EIA) and subsequent work of creating an Enterprise Metadata Management (EMM) to get to Enterprise Information Management (EIM) and related discipline processes flows from foundation Information architecture activities. The work of many groups and disciplines are in conflict because of the differences in them that stem from the lack of the foundational work of Enterprise Information Ontology. According to Wikipedia, enterprise ontologies and its engineering "aims at making explicit the knowledge contained within software applications, and within enterprises and business procedures for a particular domain." An ontology is precise approach to the structuring knowledge of a particular domain by way of an approach known as an ontology that describes its domain, the entities of that domain, and the relationships between the entities. It uses a specific methodology and documentation approach to describe the relationships among entities such as whether the relationships between terms are conceptually equivalent or partial in their semantic meaning. Ontologies also allow for the dynamic relationship linking between terms between different domains so that terms may be related to each other in a discovery based approach. Using the approach allows for the ability to have terms between silos of information to relate to other silos by way of having a central domain ontology vocabulatory on which they are based. Ontologies allow for the central mapping of semantic terms to other alternative expressions of the same domain of knowledge expressed in the form of a taxonomy or between separate domains of semantic terms in other taxonomies.
Figure: Semantic spectrums of an Ontology and the time/budge required for its development
The use of an ontology for foundational EIA work lies in the ability of the information architect to structure of the knowledge domain of an enterprise into a set of common terms and relate multiple renditions of those tribal terms to the central term. By documenting the relationships semantically in forms of equivalence part/whole or overlapping distinct terms, an improved understanding of the information technology work that emanates from the mapping can occur. The possibility and/or automated form of ETL is within reach when the relationships between domains are documented between islands of information. Isn't that what happens when ETL jobs are created between information silos today except it is done on the fly without the big picture view of a central semantic set of terms for the entire enterprise. As an example, an ETL architect will have to map between information silos and describe by way of the ETL tool, the relatioship and the exact programmatic relationship between those term meanings. Name in silo A is lastnm, firstnm, mi in silo B and they are equivalent while in silo c, name is first name in silo A. What this amounts to is that an ETL architect will eventually map the semantic meaning between terms and express the precise operation to convert between the two. In an ontology, the relationship and precision of terms can be described to an excruiciating level of detail and we can allow a futuristic ETL construct the operational precision to convert between them (e.g. ETL by way of ontology). What that means is that by defining a central semantic meaning and relationship of terms then the work of modeling the information architecture, the defining of the metadata management of terms can be documented for dynamic lookup, and construct the precise operations that are used in Enterprise Information Management. Perhaps, a top down approach to information can help us structure and organize the work all the way down into the details.
No comments:
Post a Comment