Representation options based on need
The range and variety of contexts in which signs can be used in our daily lives presents some modelling challenges. How do I capture the information I need relating to signs based on the model patterns that are present in HQDM? These are information requirement trade-off decisions against the cost and complexity of implementing the available patterns.
The starting point is to recognise that we are very unlikely to need all signs, their composition and class membership to be captured for all actual, possible and abstract things in our domain of interest. Key signs are likely to be the identifiers and names given to particular things so that they can be recognised and referred to in documents, have physical labels that carry a sign attached to them or allow people to find and understand the context of an item (e.g. a document) by its sign
.
This section presents a range of options that the Author explored while creating the examples. Not all are implemented but the aim is to illustrate what decisions could be made and the trade-offs involved. This list of options is not exhaustive. However, they are presented around information management Cases that may help you decide what approach to take for your modelling needs.
Case 1: Generic
I require a solid, repeatable implementation of all signs in my dataset handled in the same manner no matter their context
- Once implemented, it is a natural pattern (actual signs representing actual things).
- Closest to the conceptual HQDM model.
- Worth understanding before adopting any of the other patterns.
- Convenient for associating with external references to the actual signs (e.g. fabrication drawings, photos, strings, etc).
language_community
part useful for recording when changes in agreed terms, identifiers, etc change within an organisation or wider community.
- Verbose
- Can be excessive for many actual information requirements.
- Still requires careful
class
management.
Case 2: Generic without language community
I require a solid, repeatable implementation of all signs but don't need to manage changes to signs within the communities that use them
- As above, without the
language_community
part.
- As above, for Case 1.
Case 3: Generic identifiers
I require a solid, repeatable implementation of all signs for identifiers of particular (physical) things that the communities my information serves
Basic 'full' representation_by_pattern
The ellipses outside of the space-time axes denote SETs (subtypes of HQDM class
).
- Additional framework for the subtypes of HQDM
class
to support signs and their patterns. - Identifiers are commonly required in information management systems to represent assets, components, messages and packages that are sent and received, items being stocked, offered for sale and being used, etc.
- These identifiers are often specified to follow a certain pattern. Specifying that pattern, or the individual elements of the pattern that are used to compose sign instances within the model is very useful.
- Verbose.
Case 4: Identifiers for kinds
I require a solid, repeatable implementation of all signs for identifiers of the kinds of things that my information system needs records of
Basic 'full' representation_by_pattern
for a kind
The ellipses outside of the space-time axes denote SETs (subtypes of HQDM class
).
- As above, for Case 3.
- Compatible with fellow subtypes of
representation_by_pattern
likedefinition
anddescription
, often required to capture the intended interpretation / application of therepresented
kind.
- As above, for Case 3.
Case 5: Identifiers without signs
I require a solid, repeatable implementation of all identifiers that represent the things my information system needs records of. The individual signs are less important to me
Basic 'full' representation_by_pattern
without using sign
The ellipses outside of the space-time axes denote SETs (subtypes of HQDM class
).
- Less verbose. Avoids the clutter of data objects involved in creating the spatio-temporal instances of signs used.
- Retains the use of
recognising_language_community
to commit to the context in which the identifier is recognised.
- Omits the precision of what specific instances of the signs are used to represent the things that are
represented
by theidentifier
. - Can lead to adding predicates to existing
identifier
records thatconsists_of_by_class
a specificpattern
. Alternative approach requires a new instance ofidentifier
for each use of a specificpattern
. Either way this results in a data management compromise.
Case 6: Identifiers minimal
I require a solid, repeatable implementation of all identifiers that represent the things my information system needs records of. The individual signs and the connunities that recognise them are less important to me
Basic 'full' representation_by_pattern
without using sign
or recognizing_language_community
The ellipses outside of the space-time axes denote SETs (subtypes of HQDM class
).
- Least-verbose of the Cases (although further compromises could still be made).
- Missing much of the context that the other Cases preserve (to varying degrees).