Letters to the Editor
Readers are encouraged to submit letters to Coordinates. Comments about the journal in general or about specific articles will be considered for publication. The following will not be considered: announcements of events, product promotions, and questions of the sort that can be better handled by listservs. Letters may be shortened or rewritten with the consent of the writer. If a comment on a specific article is accepted, the author of the original article will be given the opportunity to reply. Letters will be arranged chronologically. Where appropriate, they will also be linked to the relevant article.
Letters must be signed, and should include e-mail address and identifying information. Please send letters to the editor at: email@example.com.
No. 1 (April 17, 2008)To the Editor:
I enjoyed David Bosse's just-released article (Coordinates, Series B, No. 9) concerning map and atlas collecting in eighteenth century America:
No. 2 (May 7, 2008)
To the Editor:
I read with great interest Jorge A. Gonzalez’ article "Problems That Arise When Providing Geographic Coordinate Information for Cataloged Maps” (Coordinates Series B, No. 8). As a fellow maps cataloger, I would like to comment on a few points, and add some additional information that is not included or apparent in the article.
Overall, the information Mr. Gonzalez provides is timely and necessary, particularly as libraries seek to provide better and more accurate access to items in cartographic collections. I personally found the beginning of the article difficult to follow and understand, although continuing to read brought many good things to light. Let me particularly note Mr. Gonzalez’ keen understanding of the importance of, and differences between, accuracy, precision, and resolution as they relate to coordinates. As a cataloger, I already have an “insider’s view,” so I was able to work through those spots less well-written and still come away with an accurate picture of the problem(s) and intended solutions. That said, a couple of things relating to cataloging standards have changed since the publication of the article, and I will share those changes below. In addition, two other things come to mind that I want to also share more information about. First, the implications of changes in international cataloging rules (which Mr. Gonzalez touches on in his conclusion) from the long-used Anglo-American Cataloguing Rules (AACR) to the new Resource Description and Access (RDA). And, second, the problem of what to do with coordinates appearing on early maps or atlases derived from a prime meridian other than Greenwich.
First, allow me to share that as of early 2008 the data input standard for including coordinates in bibliographic records has changed, both in the Library of Congress’ MARC 21 Format for Bibliographic Level, National Level Full and Minimal Requirements or “national level standard,” and in OCLC’s input standard. As Mr. Gonzalez briefly noted, at the time he was writing his article the input standard for coordinate values was “optional,” meaning that the cataloger doing the work could chose to ignore including geographic coordinate data altogether, or input this data either from the map itself or from other resources. Today the input standard in both cases is “required if applicable” in OCLC terminology, and its parallel “mandatory if applicable” in national level-speak. Either way, this means that if there are coordinates provided on the map one is describing, these coordinates must be input into the appropriate subfields in the MARC 21 standard, fields 034 and 255. If a cataloger has a map in front of him/her that does not have coordinates printed on it, then he/she can chose to either not put the time and effort into finding appropriate coordinates from an external resource, or can take this valuable step that will help other catalogers down the line, as well as researchers and retrieval systems. I personally always make the effort to include bounding-box or central point (“x,y”) geographic coordinates in my work—both because of the critical importance of this information for our users, and because the online resources mentioned by Mr. Gonzalez in his article are easy to use. The most important of these online resources are the United States Geological Survey’s Geographic Names Information System (GNIS) database, which covers places in the United States and Antarctica; and the National Geospatial-Intelligence Agency’s GEOnet Names Server database, which covers all other “foreign” places. In addition, the Library of Congress’ Authority Files, which are always my first stop in looking at finding the appropriate authorized form of a place name and simultaneously seeing if coordinates for that place are included in the authority record, is a quick and easy resource that most of us have access to via OCLC’s Connexion website. Why take the time and effort to do this? Well, as Mr. Gonzalez noted throughout his article, geographic, or geospatial, coordinates can provide the most accurate retrieval mechanism for finding maps (and other types of materials by extension) in an online catalog or similar database, and simultaneously coordinates are far less ambiguous than textual place names as they occur in subject heading strings, titles, or any other keyword-accessible area of a bibliographic record.
What about the murkier area of providing accurate coordinates for early/historic/rare maps, in which the longitude values were derived from a prime meridian other than the commonly-accepted modern prime meridian of Greenwich? The article notes that “ The coordinate information should be provided as long as the information is present or can be estimated. Greenwich is assumed when no prime meridian is identified on the map. If a different meridian is specified on an old map, the cataloger records the coordinates in Greenwich, but may choose to give other meridians provided in the notes area of a record (Mangan 2005). " Perhaps it would be useful to provide a bit more background about the cataloging rules and their interpretation that catalogers have to follow in this instance.
As one can imagine, any map collection that includes historical maps within it will encounter individual maps where the prime meridian may be denoted as anything from London to Paris to Ferro to Philadelphia and elsewhere. Greenwich was declared the prime meridian, or “starting line” for longitude values in 1884 by popular consent at the International Meridian Conference held in Washington, D.C. Given this variability of prime meridians as noted on historical maps, the Anglo-American Cataloguing Rules that we currently use as our standard for bibliographic descriptions fails to provide clear guidance on what to do when a map is based on a prime meridian other than Greenwich. Rule 3.3D1 simply states “Express coordinates in degrees ( °), minutes ( ′), and seconds ( ″) of the sexagismal system (360° circle) taken from the Greenwich prime meridian.” Thankfully, the cataloging experts who crafted our more explicit guidelines, Cartographic Materials: a Manual of Interpretation for AACR2 (CM), gave us the information we need to handle this problem. Rule 3D in CM, under “specific cases” in the set of policies in this area states: “For early cartographic items where coordinates are distorted or inaccurate, record the present-day coordinates for that area,” and the use of Greenwich as prime meridian is reinforced in Appendix B. Therefore, by providing “modern coordinates” even for a historical map, the process of geo-referencing a portion of the Earth’s surface even over a long period of time is made possible by converting other prime meridian distances to that of Greenwich. This becomes very important when one considers the ever-growing number of websites that are capable of providing a historic timeline of a given place by overlaying modern maps onto historical ones and using one of the many powerful tools that GIS provides us.
Finally, let me touch on Resource Description and Access (RDA). This is the title of a new set of international cataloging rules forthcoming in 2009 that will replace the above-mentioned second edition of the Anglo-American Cataloguing Rules. Mr. Gonzalez notes that: “The recent library information shift has created a crisis in the establishment of standards to keep up with the changes.” I believe he is referring here to the rapid movement away from using the online catalog as a primary information discovery tool, and towards using online resources driven by Google and similar search engines, and also, just as importantly, a shift away from hardcopy materials to digital/electronic ones. I’m not so sure that this constitutes a “crisis,” but it most definitely has put catalogers in a precarious position, and thus RDA is meant to address describing the vast array of digital items, and to do so with an eye more on bibliographic relationships than on describing individual items. I must admit that, for the most part, I have not kept up with the many proposals and iterations that the RDA drafts have been through over the past two or three years, but I have been reassured that, for cartographic materials at least, there will not be great changes in our descriptive practices (one hopes!). How does this relate to the topic of geographic/geospatial coordinates as provided in bibliographic descriptions? It goes back to the standards relating to whether or not to input this critical data into the record in the first place, and I am confident that with the rise of various metadata schemes the importance of catalogers including coordinates in their work will remain high. For these reasons, I urge map catalogers everywhere to make an honest effort to include coordinates in your records, going beyond the “required if applicable” standard, based on the many reasons Mr. Gonzalez gives in his article.