Date:         Fri, 30 Jun 2006 10:05:48 -0400
Reply-To:     Subject Coordinates Discussion List <[log in to unmask]>
Sender:       Subject Coordinates Discussion List <[log in to unmask]>
From:         Joe Aufmuth <[log in to unmask]>
Subject:      Re: The proposal passed MARBI!
Comments: To: Subject Coordinates Discussion List <[log in to unmask]>
Comments: cc: [log in to unmask]
Content-Type: text/plain; charset="iso-8859-1"

Hi Colleen,

Fantastic!  I am forwarding this to members of the Cartographic Users Advisory Council for consideration by their member organizations.  Thelma Thompson from New Hampshire and I are the 2006-2007 co-chairs.

Seems we already have fields for the old system of Degrees minutes and seconds (DMS) and ordinal direction.  What we need are new indexed fields, which allow a numeric search, and which is what I and several others have suggested over the past 5 years.  Wasn't a major reason for the coordinate fields request based on development of spatial interfaces to catalogues?  If so shouldn't we lean towards being able to accomplish that task?

So here is one GIS user's vote for:

Field type: floating point
Decimal Degrees (DD)
(-) sign for south latitude and west longitude coordinates.
No leading zeros for Longitude
6 decimal points in order to calculate nearest second

It is really not that hard to convert between DMS and DD.  Hopefully, these fields will be used for more than just cartographic materials.  It is the hope of a colleague, Stephanie Haas, and a whole community of scientific users, to be able to search spatial data used in catalogued scientific publications.  Ideally any item in the catalogue that has a spatial reference would have the coordinate bounding box entered into these fields.

The positional accuracy, and the consequent number of decimals of the coordinates chosen and entered by a cataloger will depend on several factors.  The data range of these fields will certainly include small scale degree only values as well as large scale DD.  One arc-second in Florida is about 100ft if I remember correctly.  So I think we need to remember that this is part of a search tool and not get carried away with pin-point accuracy.

One question I have not heard is what will be the spheroid and datum for the coordinates chosen? WGS84? Clarke 1866? GRS1980? WGS84? NAD83? NAD27? ITRF?  Will this be a standard or vary based on the source material?  If it varies, differences in spheroid and datum will produce different search results.  In Florida there is roughly 15 meters difference between NAD83 and NAD27, but in other areas of the US it is greater.  Other spheroids and datum's will have other shift problems.  

If it must be DMS, could I put in a request that "/" separators, leading zeros and ordinal values not be used and (-) sign for south latitude and west longitude coordinates be used?  At least then we still can use the fields as integer values which might be searchable on a range of values.

When all is said and done I am sure ESRI will be able to convert whatever formats are chosen.  And other independents, with additional programming, will be able to make the conversion too.  I look forward to the testing of coordinates.

Joe Aufmuth 
Interim Head Map and Imagery Library
Spatial Information Services Unit Head 
George A. Smathers Libraries 
Government Documents 
University of Florida 
P.O. Box 117001 
Gainesville, Florida 32611-7001 
352-273-0367 
Fax: 352-392-3357 
[log in to unmask]






-----Original Message-----
From: Subject Coordinates Discussion List [mailto:[log in to unmask]]On
Behalf Of Colleen R. Cahill
Sent: Friday, June 30, 2006 7:15 AM
To: [log in to unmask]
Subject: [SUBCOOR] The proposal passed MARBI!


The proposal to add coordinates to authority records
<http://www.loc.gov/marc/marbi/2006/2006-06.html> passed and in the next
year or so, this will be available in the MARC format for
authorities.  Thanks to all who worked on this, especially Jimmie
Lundgren of the University of Florida and Rebecca Guenther of the Library
of Congress, who put a great deal of effort in forming this proposal. 
Good job!

We now face some choices, primarly in what format we want the coordinates
to appear in authority records.  This can be either degrees, minutes,
seconds or decimal degrees: both have advantages and disadvantages. Here
are a few I have thought of and am hoping for your input:

Degrees, minutes, seconds format
Pro:
-Format most often printed on maps
-Familiar to most people
-Easy to quality review
-Format most often used in bib records

Con:
-Not format used by GIS search engines


Decimal Degrees
Pro:
-Format used by GIS search engines
-Can harvest data from GIS tools 

Con:
-Not as easy to quality review
-Not as familiar a format to the average person

The reason we need to have this discussion is LC will allow coordinates in
the authority file, but only in one format.  By presenting
well-thought-out reasons for a desired format, we can persuade LC to use
the most logical choice. This is also a good time to think about any
standards or recommendations, such as what is acceptable data (i.e., is
just degrees enough data? How many numbers after the decimal in decimal
degrees?)

 I am sure there are many more pros and cons to
each of these formats.  If the cartographic community can come to a
consensus about this topic, it would speed the next step to this process,
setting up a test to load coordinates in the authority files. To
facilitate this process, I will keep track of the discussion and post
compilations.  Thank you for your help.

Colleen
 Colleen R. Cahill                   |  [log in to unmask]
 Digital Production Coordinator      |  (202)707-8540
  & Recommending Officer for         |  FAX (202)707-8531
       Science Fiction & Fantasy     |  Library of Congress
These opinions are mine, Mine, Mine! |  Washington, DC 20540-4652