home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
tsql
/
doc
/
tsql.mail
/
000168_"CRLV01::HKUCE…ORION.GOV.BC.CA_Wed Jul 14 17:29:06 1993.msg
< prev
next >
Wrap
Text File
|
1996-01-31
|
5KB
|
107 lines
I have included the news release as an attachment. If you could distribute it
on to the TSQL group that would be great.
Henry Kucera
Geomatics Unit
Ministry of Environment Lands and Parks
British Columbia
Inter: Hkucera@venus.gov.bc.ca
--------------------------------------------------------------------------------
ISO acceptance of Canadian SQL3 Multi-Media 'Spatial' Proposal
The third generation of SQL is currently being developed, under the
auspices of the International Organization for Standardization (ISO).
Scheduled for the 1996 timeframe, this version: (1) will have object
oriented capabilities, (2) will be backwardly compatible with SQL 1992
(SQL2 was officially adopted last year), and (3) will have extensions
for certain application areas, including management of spatial data.
The Geomatics Unit of the Surveys and Resources Mapping Branch, British
Columbia Ministry of Environment, Lands and Parks submitted a series of
four papers on behalf of Canada for consideration by the ISO database
language Multi-Media working group. Included in this submission was a
framework for the development of Part 3 of SQL3/MM based directly on the SAIF
standard. This proposal was accepted and will be the basis for all future ISO
work regarding spatial-temporal data management in SQL3/MM.
The thrust of our submissions was fourfold.
1 Traditional approaches to managing spatial data are inadequate
because such data is inherently complex and highly variable in
structure. This is why almost all spatial data is stored outside
of DBMSs today. Ideally, all data of interest to the user should
be able to be described, stored and accessed in the same fashion.
New technologies, based on object oriented capabilities, are
required.
2 The underlying capabilities of SQL3 must be sufficiently robust to
support the extensions proposed for SQL/MM. This means that
various kinds of aggregate structures, object capabilities and the
like must be present in the basic SQL3 infrastructure. The same
underlying constructs should be applicable to animation, spatial
and temporal phenomena, full text, etc., as well as more
traditional application areas.
3 Spatial extensions should not be tied to a cartographic (static,
2-D representational) paradigm. Volumetric and temporal data must
be able to be handled, as well as a variety of spatial and
temporal relationships. In addition to forestry data, census
data, and the like, we must address data associated with global
monitoring, satellite imagery, moving oil spills, subsurface
geological structures, etc.
4 Standardized spatial extensions should be of a generic nature,
that is, without reference to roads, census tracts, ecosystem
zones, coal seams, etc. However, users must be able to define
explicitly such real world concepts in terms of the spatial
extensions and the underlying SQL data constructs.
Our submissions were based directly on the Spatial Archive and
Interchange Format, SAIF (pronounced 'safe'), the draft standard in
Canada for modelling and transferring geomatics information. SAIF is
designed to work in a database/telecommunications environment, as well
as with simple file transfers. Acceptance at the ISO - SQL level of
our approach with SAIF is particularly gratifying.
Through the ISO process we have received constructive criticism from
the Australian Data Base Language Working Group, Oracle Corporation,
Sierra Systems, MacDonald Dettwiler and Associates, Fulcrum
Technologies, the Canadian Department of National Defense, IBM and
others. With respect to SAIF, we have also benefited from work by the
Sequoia 2000 project in Berkeley, the U.S. Army Corps of Engineers
Construction Engineering Research Laboratory, the Centre de Recherche
en Geomatique at Universite Laval in Quebec, and the Centre for
Environmental and Resource Information at the University of New
Brunswick, as well as some of the GIS vendors.
The ISO submission is currently being revised in accordance with the critiques
and ISO procedures. We are also planning on releasing SAIF 3.0, to conform
with the ISO work (significant changes to SAIF relate to: the way time is
treated, a more concise way of handling gridded structures, and the
introduction of multiple inheritance.) For current data holdings in SAIF's
binary format, the new version of SAIF will be backwardly compatible with the
2.0 version. For those interested in software development, you'll be
interested in the SAIF ToolBox currently under development. The ToolBox will
provide a class library and a set of APIs for the development of applications
(e.g. translators) based on the SAIF standard.
If you would like to receive the ISO papers or any of the SAIF
documentation, please feel free to contact either Henry Kucera
(hkucera@venus.gov.bc.ca) or me (msondheim@galaxy.gov.bc.ca) through
Internet or at the address below.
Mark Sondheim,
Head, Geomatics Unit
Surveys and Resource Mapping Branch
1802 Douglas Street
Victoria, British Columbia, Canada V8V 1X4
(FAX: (604) 356-7831)