Databases Reference
In-Depth Information
So far, so good. The problem is, snapshots have come to be known (at least in some
circles) not as snapshots at all but as
materialized views
. But they're not views! Views aren't
supposed to be materialized at all; rather, operations on views are supposed to be implemented
(as explained in the introduction to this chapter) by mapping them into suitable operations on the
relvars in terms of which they're defined.
17
Thus, “materialized view” is simply a contradiction
in terms. Worse yet, the unqualified term
view
is now often taken to mean a “materialized view”
specifically—again, at least in some circles—and so we're in danger of no longer having a good
term to mean a view in the original sense. In this topic, as I've said, I do use the term
view
in its
original sense, but be warned that it doesn't always have that meaning elsewhere.
My second point is this. As some readers might know, I've been thinking about the
problem of view updating for some time; indeed, I've been trying to get view updating “right”
for many years, and this topic, which represents my most recent thinking on the subject, is far
from being my first publication in this area. Thus, where there are discrepancies between this
topic and something I've said in an earlier publication, this topic should be taken as superseding.
Here for the record is a summary list of those earlier publications:
“File Definition and Logical Data Independence” (coauthored with Paul Hopewell), Proc.
1971 ACM SIGFIDET Workshop on Data Definition, Access, and Control, San Diego,
California (November 1971)
“Updating Views,” in
Relational Database: Selected Writings
(Addison-Wesley, 1986)
“Updating Union, Intersection, and Difference Views” and “Updating Joins and Other
Views” (both coauthored with David McGoveran),
Database Programming & Design 7,
Nos. 6 (June 1994) and 8 (August 1994); republished in
Relational Database Writings
1991-1994
(Addison-Wesley, 1995)
“View Updates,” Section 10.4 of
An
Introduction to Database Systems
(8th edition,
Addison-Wesley, 2004)
“View Updating ,” Appendix E of
Databases, Types, and the Relational Model: The Third
Manifesto
, by Hugh Darwen and myself
(3rd edition, Addison-Wesley, 2006)
17
It's true that certain products do use materialization instead of operation mapping as an implementation technique—
specifically, as a technique for implementing certain “complex” retrievals on certain “complex” views—but the fact that this is so
isn't germane to the present discussion (in fact, it's a complete red herring).
