110
ical advances, ingrained assumptions about the importance, role, and difficulty of data processing
must be reassessed.
8.2.2 Data Storage
Another special class of concerns involves the distinction between storing data sets on line during
processing versus storing them in more permanent
form off line. Traditionally, on-line storage was
very limited, so off-line storage had to be used. Unfortunately, off-line storage systems and corresponding media were expensive and very error
prone. Furthermore, conversion between on- and
off-line forms was both highly constrained and centrally controlled, that is, by the central computer
center that owned and operated this type of rare
equipment, such as nine-track tape drives. Much of
the focus in data storage was thus on the specialized techniques used at a particular site to move
data sets between on-line and off-line forms. As
with processing, storage technology has advanced
rapidly, to the point where these traditional concerns have faded in importance.
8.2.3 Data Dissemination
The third class of concerns is that of the preparation of digital data products suitable for use outside the originating organization. Traditionally,
data dissemination was more or less treated as a
data storage issue, in the sense that the key feature
of data exchange was that of producing off-line media in a portable form. This focus on producing data
on appropriate media often obscured the need to attend to the inclusion of all the data actually needed
to facilitate reuse. Also, organizations were generally slow to recognize that their own interests were
best served by maintaining their data resources in
a manner that promoted both internal and external
reuse. Data dissemination concerns all too often
have been thought of as an extra burden imposed
by external governmental or support organizations.
As media and format issues have faded in importance, the real issues in both internal and external
dissemination have become more prominent: how
does an organization manage its information resources to facilitate both internal and external use
as a standard business practice, rather than as a special, extra requirement?
Each of these classes of traditional data management issues shares a common theme; in the past,
the relevant information-processing technology
was expensive and therefore relatively scarce, with
the result that most projects operated in what was
Ecological Data Storage, Management, and Dissemination
assumed to be an information-technology-poor environment. Today's situation is the result of a relatively rapid shift to an information-technology-rich
environment; data access, storage, and processing facilities are very cheap compared to even a few years
ago. As project staff have broken free from the constraints that limited the amount of data that they
could collect, generate, and store, their information
resources have exploded in size and complexity.
Thus concern about the careful management of a
small, scarce information resource must be refocused on the effective organization and management of a large, rapidly growing resource.
However, the explosive growth in data resources
is not uniform across all types of data. For example, manually collected data sets have been and are
likely to continue to be relatively scarce, small in
spatial scope, expensive to collect, and highly variable in quality. Because these factors limit the
amount of information that can be made available,
it is fairly easy to deal with data organization and
management issues for manually collected data
sets. On the other hand, technological advances
have made remotely sensed digital data sets readily available. Similar advances have also made access to vast hardware and software resources the
norm. The availability of raw data and processing
resources provides analysts with ample opportunities to process data sets in a variety of ways to derive new data sets. Thus data collections are increasingly dominated by remotely sensed and
derived data sets. The problem in managing the
overall information resource is how to include information about the methodology used in these
types of automated data capture and derivation.
8.3 Technical Advances and
Problem Evolution
What now seems clear is that a set of interrelated
technical advances has essentially redefined the information management and access environment in
which we all live. Not surprisingly, corresponding
fundamental changes in the information environment are affecting ecological analysis. We have
moved from an environment relatively poor in information technology to one in which it is hard to
avoid becoming overwhelmed by the available information and opportunities to process it. This
change has occurred so rapidly that we have had
little time to reflect on how the underlying technological changes affect the assumptions and practices of the past. It would be easy to assume that
advances in computing power, software capabili-
ical advances, ingrained assumptions about the importance, role, and difficulty of data processing
must be reassessed.
8.2.2 Data Storage
Another special class of concerns involves the distinction between storing data sets on line during
processing versus storing them in more permanent
form off line. Traditionally, on-line storage was
very limited, so off-line storage had to be used. Unfortunately, off-line storage systems and corresponding media were expensive and very error
prone. Furthermore, conversion between on- and
off-line forms was both highly constrained and centrally controlled, that is, by the central computer
center that owned and operated this type of rare
equipment, such as nine-track tape drives. Much of
the focus in data storage was thus on the specialized techniques used at a particular site to move
data sets between on-line and off-line forms. As
with processing, storage technology has advanced
rapidly, to the point where these traditional concerns have faded in importance.
8.2.3 Data Dissemination
The third class of concerns is that of the preparation of digital data products suitable for use outside the originating organization. Traditionally,
data dissemination was more or less treated as a
data storage issue, in the sense that the key feature
of data exchange was that of producing off-line media in a portable form. This focus on producing data
on appropriate media often obscured the need to attend to the inclusion of all the data actually needed
to facilitate reuse. Also, organizations were generally slow to recognize that their own interests were
best served by maintaining their data resources in
a manner that promoted both internal and external
reuse. Data dissemination concerns all too often
have been thought of as an extra burden imposed
by external governmental or support organizations.
As media and format issues have faded in importance, the real issues in both internal and external
dissemination have become more prominent: how
does an organization manage its information resources to facilitate both internal and external use
as a standard business practice, rather than as a special, extra requirement?
Each of these classes of traditional data management issues shares a common theme; in the past,
the relevant information-processing technology
was expensive and therefore relatively scarce, with
the result that most projects operated in what was
Ecological Data Storage, Management, and Dissemination
assumed to be an information-technology-poor environment. Today's situation is the result of a relatively rapid shift to an information-technology-rich
environment; data access, storage, and processing facilities are very cheap compared to even a few years
ago. As project staff have broken free from the constraints that limited the amount of data that they
could collect, generate, and store, their information
resources have exploded in size and complexity.
Thus concern about the careful management of a
small, scarce information resource must be refocused on the effective organization and management of a large, rapidly growing resource.
However, the explosive growth in data resources
is not uniform across all types of data. For example, manually collected data sets have been and are
likely to continue to be relatively scarce, small in
spatial scope, expensive to collect, and highly variable in quality. Because these factors limit the
amount of information that can be made available,
it is fairly easy to deal with data organization and
management issues for manually collected data
sets. On the other hand, technological advances
have made remotely sensed digital data sets readily available. Similar advances have also made access to vast hardware and software resources the
norm. The availability of raw data and processing
resources provides analysts with ample opportunities to process data sets in a variety of ways to derive new data sets. Thus data collections are increasingly dominated by remotely sensed and
derived data sets. The problem in managing the
overall information resource is how to include information about the methodology used in these
types of automated data capture and derivation.
8.3 Technical Advances and
Problem Evolution
What now seems clear is that a set of interrelated
technical advances has essentially redefined the information management and access environment in
which we all live. Not surprisingly, corresponding
fundamental changes in the information environment are affecting ecological analysis. We have
moved from an environment relatively poor in information technology to one in which it is hard to
avoid becoming overwhelmed by the available information and opportunities to process it. This
change has occurred so rapidly that we have had
little time to reflect on how the underlying technological changes affect the assumptions and practices of the past. It would be easy to assume that
advances in computing power, software capabili-