Gemeinsamer Bibliotheksverbund ISBN

When I import via "Add Item(s) by Identifier" from the Gemeinsamer Bibliotheksverbund (GBV), it sometimes imports the ToC as well. While I consider this a great additional service, the ToC is automatically renamed to match the parent file. This is problematic because I then cannot tell apart whether I have the whole document or just the ToC. Is there a good way to (automatically) distinguish between both PDF attachment forms?
  • The TOC attachement should have the title "Table of Contents PDF" from the translator. Are you using ZotFile? I thinks this additional plugin might rename the attachements...

    Normally, documents with ISBN are books, do you have fulltext of the whole book attached to them? (Otherwise you could try to search for the itemType.)
  • edited May 6, 2015
    Dear zuphilip,

    You're right. When I switch off the ZotFile feature it works fine.

    One thing I've noticed is that not all titles available in GBV are imported and that not all GBV entries that have a ToC have it imported. I know that because I used ISBN from entries directly from the GVK catalogue. Do you know why that is?

    With German titles, I separate title and subtitle normally with a point ".". Only when the German titles explicitly deviate from this pattern, then I do so too. The GBV imported entries separate title and subtitle with a double point ":". Is there any way to tweak that to my preference without having to change it manually after every import?
  • One thing I've noticed is that not all titles available in GBV are imported
    Can you give an example here? (We focus on print material and exclude ebooks from GBV catalogue.)
    and that not all GBV entries that have a ToC have it imported
    Example? IMO, currently we only add TOCs from the DNB which are following a special link pattern and are therfore easy to detect.
    Is there any way to tweak that [separate title and subtitle with a point ".".] to my preference without having to change it manually after every import?
    That sounds more like a general question about the data model in zotero. @adamsmith: Any ideas or comments?
  • Yeah, as you may remember, I've been thinking about doing a better job with title/subtitle divisions, but I don't think we have any imminent solution for this and we're not going to change this in the translator, although I understand why people may want that.
  • One thing I've noticed is that not all titles available in GBV are imported

    Can you give an example here? (We focus on print material and exclude ebooks from GBV catalogue.)
    For example:

    I've searched for the following ISBN: 978-3-8274-1723-7

    The corresponding book is:

    Spitzer, M. (2007). Lernen: Gehirnforschung und die Schule des Lebens. München: Spektrum Akademischer Verlag.

    I've got a result from "Library of Congress ISBN" instead of "Gemeinsamer Bibliotheksverbund ISBN".

    When searching in http://gso.gbv.de/DB=2.1/ with the ISBN, I get plenty of search results. I therefore wonder why GBV data is not imported. The GBV entry links to the DNB ToC http://d-nb.info/980577039/04 too.

    (I'm sorry for not replying more quickly. I'm very occupied by my work at the moment.)
  • We default to library of Congress as our first choice. That's on purpose and I'm not sure we'll want to add user preferences to the order in which we go through databases.
  • If an ISBN is available from library of congress, Zotero will prefer that over other sources. Is there an issue with LoC metadata?
  • There are no big issues with the LoC metadata with the example above.

    LoC:
    Spitzer, M. (2007). Lernen: Gehirnforschung und die Schule des Lebens (1. Aufl.). München: Spektrum Akademischer Verlag.
    The bits that I have to do manually are:
    1) Separating title and subtitle with a point instead of a double point. This is a convention at least with the German titles in the field of education.
    2) I also remove the "1. Aufl." because when I quote a German title in an English text, I want it to be "1st ed." So if it can be done, it would probably be better just to import the number (i.e. "1") to highlight the respective edition. I usually only keep information on the edition if greater than one.

    After my manual modifications the reference looks like this:
    Spitzer, M. (2007). Lernen. Gehirnforschung und die Schule des Lebens. München: Spektrum Akademischer Verlag.
    The great thing about GBV data is that it imports ToCs as well (if available at the DNB). There seems to be no equivalent feature for the LoC. While this is not a mayor issue, it's still a feature that I value.
  • edited May 22, 2015
    Dear zuphilip,

    Is it also possible to attach GBV ToCs?

    For example for the following entry, where no DNB ToC is attached?

    Höhne, T. (2015). Ökonomisierung und Bildung. Zu den Formen ökonomischer Rationalisierung im Feld der Bildung (essentials). Wiesbaden: Springer VS.
  • In principle, this is possible. However, in the MARC format there is a general URL field (no. 856) and I am not sure if it is desirable to import every URL (maybe not) or how to restrict to URLs to TOCs optimally. E.g.

    datafield tag="856" ind1="4" ind2="2">
    <subfield code="u">http://deposit.d-nb.de/cgi-bin/dokserv?id=5181368&prov=M&dok_var=1&dok_ext=htm</subfield>;
    <subfield code="m">Verlag: MVB</subfield>
    <subfield code="3">Inhaltstext</subfield>
    </datafield>
    <datafield tag="856" ind1="4" ind2="2">
    <subfield code="y">Inhaltsverzeichnis</subfield>
    <subfield code="u">http://www.gbv.de/dms/zbw/821525689.pdf</subfield>;
    <subfield code="m">V:DE-601;DE-206</subfield>
    <subfield code="3">Inhaltsverzeichnis</subfield>
    </datafield>
  • I might be a bit naive but would the following field not signify that it is a ToC to be imported?

    <subfield code="3">Inhaltsverzeichnis</subfield>

    or this one

    <subfield code="y">Inhaltsverzeichnis</subfield>
  • AFAIK there is no subfield code, which specify the ToC. When cataloguing we don't insert "dedicated ToC URLs". The 856 field can contain different kinds of URLs.
  • I see, that is good to know. The suggestion by @zurpher was to take the url from the field 856, if "Inhaltsverzeichnis" is written in subfield 3 or y. This sounds reasonable. Do you know if the text in the subfields is always (or at least often enough) exactly "Inhaltsverzeichnis" for GBV entries?
  • Cataloguers can provide the link text "Inhaltsverzeichnis", but its not mandatory. But most of the links in 856 (in pica+ its: 009Q) were inserted by routines, i guess. Therefore there is reason to assume that the text is in most cases exactly "Inhaltsverzeichnis".

    According to the concordance from DNB http://d-nb.info/103016455X/34 (pdf, p.134) the link text should be in subfield y. I don't know why it's in subfield 3, too.
  • I did a quick scan of the values in 856 (aka 4099 aka 009Q) in the GBV catalog. Is the link text "Inhaltsverzeichnis" available?
    GBV-ToC: y:yes, 3:yes
    DNB-ToC: y:no, 3:yes
    Dandelon: y:no, 3:yes (Example: http://gso.gbv.de/DB=2.1/PPNSET?PPN=599705590)

    I think these 3 are the most important sources for ToC. Subfield 3 is not visible in the catalog frontend, but it seems to be a more reliable indication than y. My conclusion: check for both subfields.
  • Okay, thank you for the analysis. I will try to work this out later in the translator and create a pull request for the official repo. This may take some time...
  • The GBV translator is now updated and should save more Table of Contents, i.e. whenever "Inhaltsverzeichnis" is written in a subfield of marc field 856. Enjoy!
  • Great job! Thank you zuphilip and phu.
  • Just updated to 4.0.28.1 and tried it out with the following ISBNs

    978-1-4073-0412-0
    978-3-8274-1723-7
    978-3-658-08973-3

    but no luck, there are no ToC links after the import. Is it just me?
  • The first two examples are by standard handled by the Library of Congress ISBN translator (check the field "Bibliothekskatalog"), the last one is working for me and gives me also a TOC.
  • "Bibliothekskatalog" says "Gemeinsamer Bibliotheksverbund ISBN" for all 3 examples. I'll check the new Zotero version on another computer and come back later.
  • Everything is fine on my private computer. Thank you, Philipp!
  • edited November 25, 2016
    When I imported data of two new books via "Add Item(s) by Identifier" from the Gemeinsamer Bibliotheksverbund (GBV), not only the editors but also the publishers have been listed as editors. Is there a way to avoid that?

    Here are the ISBNs that I have used:

    978-3-7815-2123-0
    978-3-658-15411-0

    EDIT:

    I experienced the same issue now with Open WorldCat and the following ISBN.

    978-3-8031-2765-5
  • Another point I have noticed is that when I import data via the "Gemeinsamer Bibliotheksverbund ISBN" recently, the Language field populated with an "und" instead of e.g. "de", "de-DE", "German" or "Deutsch" as it should for publications in German.

    Example ISBN: 9783840927638
  • Okay, you are right on all of these issues, but the reason and fix might be different:

    a) publishers as editors in GBV: The reason is that this relation is now also saved in MARC field 710, but for the authors/contributors this has to be skipped. I will try to fix that on the Zotero side.

    b) publishers as editors in Open WorldCat: This seems to come directly from their RIS and is wrong on their side. Someone should tell them...

    c) wrong language code "und" in GBV: This is in their MARC field 041 and seems to be an error they should be able to correct. I will let them know.

    @zurpher Thank you for the reports here. This is valuable for metadata quality in Zotero translators. It sometimes takes a little longer until I can look at the issues.
Sign In or Register to comment.