Dear Petr,



I’m aware of the technologies you are developing to make geodata-searches easier. And with envy I sometimes look at the Swiss situation. I’ve recommended http://boundingbox.klokantech.com/ to my Dutch colleagues and some start to use it as in former times they usually were loath to add geographical coordinates to the cataloguing-database. For most it didn’t make sense as they could not be used in a search-engine. This may probably be one of the reasons why I sometimes get the answer, “but we already state all possible geographic data and related data in the verbal description”. At the Koninklijke Bibliotheek we added geographical coordinates to all our description because we believed in the idea propounded by Dr. Hans van der Waal (Unfortunately I cannot find the literature reference for his article, I think from 1974) that these coordinates could be used to find relevant documents in visual displays or in special queries.



The question I recently put forward, however, is,



A.    are researchers, who use geo-data, needing a user-interface with coded data to help them find the necessary documents/data



B.     are we willing to spent time in encoding our bibliographic data into fixed field positions in MARC 21 worldwide



C.     can we convince OCLC to design user-interfaces in such a way that these coded data can be used in queries.



I hope that we can have a fruitfull discussion concerning these subjects.



With kind regards,



Jan



Van: Maps-L: Map Librarians, etc. [mailto:[log in to unmask]] Namens Klokan Petr Pridal

Verzonden: woensdag 24 februari 2016 18:34

Aan: [log in to unmask]

Onderwerp: Re: MARC 21 coded data for cartographic/geo-documents



Dear Jan,



The Swiss libraries have a unique geographical search system for union metadata map catalog - allowing extremely user-friendly and fast search for queries like you mentioned: “Do you have a map covering that area with that subject/theme on that scale and/or around that time?”



See and try: http://search.kartenportal.ch/



It has also support for map series and visualisation of the indexes of the sheets - just zoom out on this view: http://search.kartenportal.ch/#series&id=33782326X



It is indexing over 335.000 metadata records in MARC format - collected by the librarians for last decades. Indexing is automated - modified metadata records are overnight harvested via OAI-PMH from the SwissBib union catalog. Exciting is also the unification of the subject/theme across the 4 languages used in Switzerland.



To simplify the generating of metadata records we have made a basic free online tool for map librarians: http://boundingbox.klokantech.com/.



This tool prints out the MARC 034 fields in correct formatting for a visually selected region - ready for copy&paste to Aleph client and other cataloguing software tools. This tool has been described by a Harvard map librarian in and article in MAGIRT BaseLine magazine linked from http://blog.klokantech.com/2013/06/bounding-box-tool-for.html - and the online tool is in use everyday by many map librarians in Europe and USA every day.



The search technology and user interface behind the system is called MapRank. It is product which any library can use for indexing their own metadata to create a geographical search page integrated with their own websites or OPAC. As the main developer of this technology I would be personally very keen to work on new applications of this search system for map librarians in Netherlands or elsewhere.



Feel free to contact me at: [log in to unmask]<mailto:[log in to unmask]> if you have any questions.



Best regards,



Petr Pridal, Ph.D.

Klokan Technologies GmbH

http://www.klokantech.com/



P.S.

You may know the search system from the popular project http://www.oldmapsonline.org/ which is similarly indexing scanned zoomable maps available online. It comes with a mobile app for iOS and Android - applicable for a library catalogs as well.



On Tue, Feb 16, 2016 at 8:53 AM, Jan Smits <[log in to unmask]<mailto:[log in to unmask]>> wrote:

[cid:image001.gif@01D16FE1.10017230]

Coded data for cartographic bibliographic descriptions

Many a time in my former life as a map librarian, patrons, instead of requesting a specific cartographic document, would usually posed questions like: “Do you have a map covering that area with that subject/theme on that scale and/or around that time?” And some were even more specific and requested that the data on the cartographic material were depicted using a certain technology or created from certain sources, especially with remote sensing maps. They viewed our collection more as a collection of information than as a collection of objects.

After an interview of between 10 and 30 minutes we were pretty sure what the patron wanted and could proceed or not to produce the document(s)  needed.



When the DUMC (Dutch Union Map Catalogue)-project was started in the early 1970s under the guidance of professor Cornelis Koeman (who was first a geodetic engineer and later became the first professor in cartography in The Netherlands) there was the idea of how a bibliographic apparatus might be able to answer such questions. The idea was that one should not only describe the cartographic materials at hand bibliographically but also encode (i.e., use codes for) as much possible data regarding the content of the document on fixed positions in certain fields. When all these coded data were indexed it should be possible to build a user interface that, when queried, could come up with answers that approximated the patron’s request.



When I started to work with the DUMC in 1979 the possibility to add coded data was realized in Unimarc, the system then in use. Thus happily we added as much as possible codes to our descriptions. Till 1984 they included even geographical coordinates that were converted from the verbal coordinate-field (255$c) to positive coordinates (decimal degree format) in the 034 field as computers only could do arithmetics with positive values. Unfortunately this feature disappeared when in 1983 we had to convert our database to the format used by Pica, the foundation that develops and administers the bibliographic database we use. Pica for some years now is incorporated into OCLC.



Unfortunately the creation of a user interface that could handle all these encoded data points was never realized. This was amongst other things the result of another conversion to a new DUMC-system in 1989 and a new conversion to Pica in 2009. The DUMC died a silent death in 1999 as the participants did not believe that a separate bibliographic system for cartographic materials could survive in a media world that was striving for one-stop-shop systems for all library materials.



However, up till now I have been adding coded data to all cartographic descriptions, now numbering some 30,000 items. Even my colleagues at the university libraries are starting to do so more and more, though not as fanatical as I would like.



MARC21

Though not very far away from my  retirement I’m suddenly, like all my colleagues, confronted with the possible conversion of all bibliographic data from our present Pica-MARC to MARC21. To make the conversion as smooth as possible I  have been requested by OCLC|Pica to evaluate whether coded and verbal data concerning cartographic materials will be given their right place in the MARC21 fields. After struggling to understand where certain coded data belonged I was left with some unresolved matters. After e-mail correspondence with Paige Andrew (whom I also thank for editing this document) I was able to recognize more possible fields in which to place coded data. However, not all data we encode can be put in the MARC21-format. Although much concerning the content or character of a  map can be noted in the 500-fields, a lot of  information is not described using codes.

Looking through the 006/008 fixed-fields we see that there is the possibility for other cataloguers to fill in coded data, such as for ‘all materials’, ‘books’, computer files’, ‘music’, continuing resources’, ‘visual materials’ and ‘mixed materials’, and in 007 several more.



While corresponding with Paige I have come up with the question: “Are the coded data being used one way or another by library-patrons or are we just filling up space? And if yes, do we enter enough coded data in our descriptions?”



I would like to discuss this with colleagues world-wide. Of course the first question is of most importance (i.e., are patrons able to use our retrieval systems based on coded data), before we can discuss the next one (i.e., are map catalogers including enough coded data to make different forms of retrieval possible/effective). But to help participants with context I’ll list next where encoded information can be found in MARC21 and what data, in my opinion, is lacking.



(when discussing fixed field codes we of course have to keep in mind that the cartographic/geo-field is constantly creating new expressions, that makes the fields similar to ongoing resources)



006 - Fixed-Length Data Elements-Additional Material Characteristics (R): Maps

007 - Physical Description Fixed Field-General Information (R): Map (007/00=a)<https://www.loc.gov/marc/bibliographic/bd007a.html>, Globe (007/00=d)<https://www.loc.gov/marc/bibliographic/bd007d.html>

008 - Fixed-Length Data Elements-General Information (NR): Maps

034 - Coded Cartographic Mathematical Data (R)

045 - Time Period of Content (NR)



In 1997 coded data concerning prime meridians became obsolete in MARC21. These are important for most pre-1884 maps and maps, e.g., from France and Spain, that use the Paris or Madrid meridian.

Up till that time probably the same codes were used as by us:

aa



Greenwich



International prime meridian



ab



Amsterdam



E 4 53 01



ac



Athene



E 23 42 59



ad



Batavia



E 106 48 28



ae



Bern



E 7 26 22



af



Bogota



W 74 04 53



ag



Bombay



E 72 48 55



ah



Brussel



E 4 22 06



ai



Cadiz



W 6 17 42



aj



Kaapstad



E 18 48 41



ak



Caracas



W 66 55 50



al



Kopenhagen



E 12 34 40



am



Cordoba



W 64 12 03



an



Ferro



W 17 39 46



aE



Helsinki



E 24 57 17



ap



Istanbul



E 28 58 50



aq



Julianehaab



W 46 02 22



ar



Lissabon



W 9 11 10



as



Londen



E 0 05 43



at



Madras



E 80 14 50



ba



Madrid



W 3 41 15



bb



Mexico city



W 99 11 40



bc



Moskou



E 37 34 15



bd



München



E 11 36 32



be



Napels



E 14 15 42



bf



Oslo



E 10 43 23



bg



Parijs



E 2 20 14



bh



Peking



E 116 28 10



bi



Philadelphia



W 75 08 55



bj



Pulkovo (Leningrad)



E 30 19 39



bk



Rio de Janeiro



W 43 01 21



bl



Rome



E 12 29 05



bm



Santiago (Chili)



W 70 41 00



bn



Stockholm



E 18 03 30



bE



Sydney



E 151 12 23



bp



Tirana



E 19 46 45



bq



Tokyo



E 139 44 40



br



Washington, DC



W 77 00 34



bs



New York City



W 74 00 29



xx



other



yy



no prime meridian



zz



unknown





Cartographic imaging technology definitions/descriptions from cataloging/other reference sources. (terms derived from: KW = Dutch cartographic dictionary, CM = Cartographic Materials: A Manual of Interpretation for AACR2, Second Edition, BDI = Dutch library dictionary)



Anaglyps (KW, CM)



Polarized map (KW)



Planimetric map (KW)



Diagram map (KW, CM)



Flow map (KW)



Dot map (KW)



Diagrammatic map  (KW)



Choropleth map (KW)



Dot map (KW)



Dasymetric map (KW)



Grid map (KW)



Isoline map / Isopleth map (KW)



Anamorphic map (BDI, CM)







Stereoscopic model / statistical block digram (KW)



Mental map (KW)



Elevational drawing with horizon (incl. bird’s eye view and pictorial relief map) (KW)



Elevational drawing without horizon (incl. bird’s eye view and pictorial relief map) (KW)



Mapview (in combination with map in orthographic projection and details which are depicted in elevation)



Pictomap (KW)



Random dot map



Grid map (KW)



Map without grid





Form of the cartographic document, like:



Atlas (KW)



Atlas-map



Map (KW)



Wall map (KW)







Chart (KW)



Town map (KW)



Cadastral plan (KW)



Architectural drawing or plan



Picture postcard (BDI)



Map on calender



Map as puzzle



Map as kit



Index / Sheet index (KW)





(sketch map, -, route map,



Topographic elevational drawing (KW)



Gore map (KW)



Lineprinter map (KW)



Block diagram (KW)



Profile (KW)



Architectural rendering (CM)



Globe (KW)



Braille [map] (BDI)



Facsimile map (KW)



Uncontrolled photomap (CM)



Controlled photomap (CM)



Photomosaic (KW)



Photomap (BDI)



Orthophotomap (CM)





Again, my aim for posting this on MAPS-L is to start a conversation amongst catalogers in particular, but more broadly amongst anyone with an interest in seeing changes to our cartographic resources cataloging based on possibly using codes more frequently for describing data points in tandem with textual data. If there is a relevant need for adding to or changing the MARC21 descriptive cataloging standard to enhance the user experience in terms of retrieving needed maps (or similar) through assisting computers by the use of more codes then how do we move forward to accomplish this?



With kind regards,



Jan Smits

Senior cataloger/Senior information specialist

Koninklijke Bibliotheek, National Library of The Netherlands









[Koninklijke Bibliotheek, National Library of the Netherlands]











Prins Willem-Alexanderhof 5 | 2595 BE Den Haag

Postbus 90407 | 2509 LK Den Haag | (070) 314 09 11 | www.kb.nl<http://www.kb.nl/>











[http://www.kb.nl/sites/default/files/dots.jpg]











English version<http://www.kb.nl/en/email> | Disclaimer<http://www.kb.nl/disclaimer>













--

http://blog.klokan.cz/

http://www.maptiler.org/

http://www.oldmapsonline.org/