another FYI post. Dan




Peter McKeague <[log in to unmask]> on 10/03/2001 08:42:52 AM

Please respond to "The Forum for Information Standards in Heritage (FISH)"
      <[log in to unmask]>



 To:      [log in to unmask]

 cc:      (bcc: Dan Page/HullOttawa/PCH/CA)



 Subject: symbologies versus underlying data standards






The GIS is an incredibly flexible tool.  It enables all sorts of data (gathered
for all sorts of reasons, to all sorts of undefined standards) to be brought
together, shared, analysed and abused.  How we wish to see our data displayed
may not be appropriate if that data is displayed against third party data with
colour and symbology conflicts that may arise.

Moving towards standardising the cartographic representation of graphic elements
(linework, point symbology, area features) within a GIS (such symbology exists
for geological mapping) is a presentation issue.  With style sheets or legends
in a GIS, all you are really standardising is the representation of the
underlying data.  Legends do not address fundamental issues such as the
interpretation or quality of the information.  It is with this data, where
standards should first be defined.

What we really should be asking is how much standardisation should exist between
datasets commissioned from external sources.  Are they simply a digital record
of a particular event, be it an excavation or a field survey, associated with a
site record?  Alternatively we could apply controlled terminology to information
describing the character of the event so that the limit of excavation (as
opposed to trench edge, or edge of trench etc.) is used consistently between
contractors.  This would enable key information to be plotted systematically in
the GIS but not affect how the detailed archive was structured within the
project.  Or should we be more ambitious and demand the ability to search across
datasets?  If digital archives are organised systematically, the GIS could
provide detailed spatial analysis of excavation archives enabling the user to
retrieve for instance all post-holes that produced flint from prehistoric
excavations.


We should address issues concerning input of data including orginal data capture
standards including documentation (metadata) accuracy and precision, data
migration from CAD to GIS and between GIS systems.


How do we share data?   Current supply formats include industry standard shape
files and MapInfo export formats, dxf  and various types of database.  Future
developments will impact on the GIS industry.  Already we have GML on the
horizon and as database developers like Oracle become involved,   exchange
mechanisms to serve the wider GIS community will inevitably adapt to meet their
demands.

There are also longer term issues concerning how we store and maintain our data
which need to be addressed (though perhaps not by this discussion forum).


Peter McKeague
RCAHMS



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
RCAHMS
(The Royal Commission on the Ancient and Historical Monuments of Scotland)
John Sinclair House, 16 Bernard Terrace, Edinburgh, EH8 9NX
Tel: 0131 662 1456
Website: www.rcahms.gov.uk
****************************************************
This email (and any files transmitted with it) is
confidential and intended solely for the use of the
individual or entity to whom it is addressed. It has
been swept for the presence of computer viruses.
****************************************************