3.1 Technical Issues
3.1.1 Metadata
Standards
ISO 19115 is the
schema required for describing geographic information and
services. It provides information regarding identification,
extent, data quality, spatial and temporal schema, spatial
reference, and distribution of digital geographical data. It is
applicable to the cataloguing of datasets, clearinghouses and
offers a wide range of geographical applications. The ISO 19115
metadata standard reflects other standards such as FGDC, TC 287,
and ANZLIC (New Zealand Standards Concern) and from other
inputs. It includes catalogue elements and structures, provides
additional detail in excess of FGDC metadata, including special
sets of raster and imagery information. 19115 consist of the
same elements as the FGDC but most of the element names have
been modified using international standard terms. However,
similar as it may be to FGDC, the ISO make-up is largely
conditional with many optional elements and structures. It
includes approximately 50 fields which covers the who, what,
when, where, why, and how questions of the datasets better than
FGDC but it may not be adequate for detailed description
requirements. Despite all of this, ISO 19115 has yet to be
published, for it is in its final stages of being approved as an
International Standard.
There are
approximately 50 fields of possible information for creating
metadata; however they may not all be fully populated by data
producers due to the lack of information. Hence, there should
be a minimum number of elements that should be maintained for a
particular dataset; these are known as core metadata elements.
There are the core metadata elements required for describing a
dataset (Table 1).
“M”
means it is mandatory,
“C”
means it is conditional, and
“O”
means it is optional. Similar to FGDC, some elements are
required by the data producer, the producer must complete the
conditional elements if information is available and the
optional elements are discretionary.
The FGDC metadata
structure is the standard Canada is currently using for
regulating their metadata; however, FGDC is an American standard
that was created on the order by President William Clinton in
1994.
The individual
metadata elements in the FGDC metadata standard are made up of
compound elements and normal elements. A compound element is a
group of data elements and other compound elements. They are
described by data elements, either directly or through
intermediate compound elements. Compound elements represent
higher-level concepts that cannot be represented by individual
data elements. Data elements are a logically primitive item of
data and FGDC consists of seven compound elements. These
elements can be further classed into three classes: mandatory,
mandatory-if-applicable, or optional. Mandatory elements must
be provided; mandatory-if-applicable must be provided if dataset
exhibits defined characteristics. In addition, production rules
assist the make up of the standard by specifying the
relationship between compound elements and data elements and
compound elements.
One of
the foremost differences between the FGDC and ISO 19115
standards is that FGDC is an American Standard and is not
internationally recognized by other countries. As a result, ISO
TC 211 is in the process of publishing their version of the
metadata standard that will be used worldwide by other
countries, which will help minimize issues of interoperability
between countries when sharing data. Another difference is that
the ISO 19115 contains more elements, specifically more optional
elements to address many known deficiencies in the FGDC. New
terminology is also introduced in the ISO 19115 because of
international influences to define terms clearer. As mentioned
above, their structures are quite diverse, FGDC has a
hierarchical structure, and however, the ISO 19115 has an
object-oriented model. On the other hand, these two standards
have similarities as well. To state the obvious they are
standards, which are used to describe data. They include
similar elements/packages like identification information, data
quality information; reference systems contact information and
more. The difference is that the FGDC is not ISO compliant and
that is why they are revising the FGDC standard to harmonize
with ISO.
3.1.2
The application of eXtensible Mark-up
Language
XML is designed to
organize and manipulate information independent of what
operating system it is running on and thus, suggests the
possibility of true standardization and interoperability of the
metadata in GIS. The primary reason for this is that it is an
extremely useful language for storing content that requires
constant updates because it separates content from presentation.
XML files have no presentation information stored in them like
HTML files do. This leaves the files themselves much easier to
read, modify and create. Another major advantage is that XML
allows for the creation of object-orientated databases (OODB),
which are very useful for spatial data dictionaries and
metadata. Once an XML template is created, i.e. a list of all
the different XML tags sometimes ranging upwards of 100 or more
different tags, then hundreds or even thousands of objects can
be catalogued, each represented by one XML file. The best
analogy for this system is an old-fashioned library
card-catalogue system: each card in the catalogue had a list of
elements (e.g. name, ISBN, date published, etc.) and each
different book would have unique values for those elements. The
fact that one style sheet can be used to not only view, but
modify many XML files has a great advantage over other online
database formats. All that is required is an embedded script
that will draw out the desired elements from the XML file, and
disregard unneeded elements. In the case of the SFU-SDD, an
embedded java script within an HTML file associated with the
ESRI Metadata Explorer is literally a metadata generator
A spatial data
dictionary designed in an Object-Oriented Database (OODB) format
has an immense amount of potential for next-generation search
engines, opening the door for context to be included in the
search engine’s
processing power. It would be possible to implement ontologies
and semantics with XML in such a way not possible with
traditional metadata storage. As mentioned above, XML is
concerned with storage of content, but not all of the content
needs to be metadata. A secondary set of tags could be embedded
in the XML files to serve as the basis for artificial ontologies
within the dataset. The definition of XML codes used on a web
page can also be used to build ontologies.
XML allows users to add arbitrary
structure to their documents, but does not describe the meaning
of the structures. This leads to the fact that XML cannot
directly be used to represent the conceptual database schemes or
the associated ontologies; thus, the World Wide Web Consortium
has developed the Resource Description Framework (RDF)). The
RDF scheme can use its own internal domain ontology, or it can
link to other known sites around the internet, and incorporate
their domain ontologies for a specific feature as well.
|
