Incomplete COinS import from German Education Index

zurpher
When importing COinS from the German Education Index to Zotero, the abstract and the tags, i.e. Index Terms, are not imported. Abstract and tags are imported via COinS when using Citavi.

Do you know why not all the available data is imported by Zotero? Is there a way to fix this?
«1
  • zuphilip
    I think "rft.subject" and "rft.description" are not part of the COinS standard. At least this would make sense, because the standard was originally invented for a linking system (OpenURL), where you do not need annotations. Can someone confirm that?
  • noksagt
    Correct. ANSI/NISO did not enumerate either of those in the OpenURL 1.0 specification (pdf).

    Those are from Dublin Core, so aren't valid technically for the rft_val_fmt chosen.

    But there is also probably little ambiguity about what is meant, so I'd probably be fine if we imported from those keys (assuming the code was commented to denote it was non-standard and that we didn't export them).
  • aurimas
    +1 from me
  • adamsmith
    agree; don't see why not.
  • zurpher
    Great news. Thanks for your swift responses.

    @zuphilip

    Can you let me know when these modifications are live. Thank you!
  • zuphilip
    Okay, let us see if and how we can implement this. I try to keep you updated here, but it may take a little longer, since this function is more deeply in the Zotero core code. @zurpher: Feel free to ping again here after some time.
  • zurpher
    @zuphilip

    Did you find some time to look into this?
  • adamsmith
    yes, this will be in the next version of Zotero. zuphilip's proposed changes were already accepted. COinS changes aren't part of the regular translators (which update daily), so they always take longer to go live.
  • zurpher
    Thank you for the update. I'm looking forward to this. It's going to be very helpful to have the summary and tags imported automatically.

    Do you happen to know when the next version of Zotero will be released?
  • adamsmith
    edited November 24, 2015
    Hopefully by December 15...
  • zurpher
    The abstract is now being imported. Great.

    The tags are imported too but all in only one field:

    For example, the following keywords are only 1 tag instead of 16 seperate ones:
    Empirische Untersuchung; Kommunikation; Verhalten; Ethnografie; Schulorganisation; Schulklasse; Versuchsschule; Lehrer; Schüler; Interaktion; Gleichaltrigengruppe; Unterricht; Norm; Ordnung; Forschungsprojekt; Bielefeld
  • noksagt
    There is nothin in the standard that defines a delimiter for keywords. I'd imagine repeating the tag would be preferred (this is the preferred handling for the author tag). I'd suggest contacting the server to ask them to change it.

    I have no strong opinion on whether we should be splitting on delimiters of our choosing for poorly formatted COinS (we occasionally do this for e.g. poorly formatted RIS). However, adding this would impact other sites and it is unclear to me that is what we'd always want to do and we should give it some consideration before adding this based on one side that uses non standard COinS.
  • adamsmith
    I think I'd be OK to split tags on semicolons. Who puts semicolons in a Keyword? But will wait for zuphilip, he's the real librarian and gets to be the authority on openURL questions ;)
  • zuphilip
    As mentioned above also "rft.description" is not part of the COinS/OpenURL standard but it is part of Dublin Core metadata format. Thus, technically these fields aren't valid for "rft_val_fmt=info:ofi/fmt:kev:mtx:book" as it is choosen in the database. Moreover, it is not possible to mix different namespaces as mentioned here http://epub.mimas.ac.uk/openurl/KEV_Guidelines-200706.html (Example 5.10).

    I think that simple Dublin Core format would prefer to have multiple subject fields rather than one field containing a list of subjects. I don't know how other websides are handling this.

    Maybe, the easiest would be to build a simple translator for the specific database which would then basically just calling the COinS translator. But we could then restrict the splitting of the tags to this database. Moreover, it would be possible to handle multiples.
  • zurpher
    Okay, as I understand it, zuphilip will try to implement a fix that applies to FIS Bildung (German Education Index) alone. I therefore do not have to contact the provider to see whether they can implement changes on their side.

    Someone from the German Education Index suggested to me that import into Citavi works fine. Maybe it is possible to find out which solution they have implemented.

    ps: Happy New Year to you all.
  • zuphilip
    Yes, that is the plan. Thanks for the reminder (ping me again if you don't hear after some time here).
  • zuphilip
    @zurpher: A new translator for "Fachportal Pädagogik" is now up. It should being added during an automatic update, or alternatively you can update manually your translators. Let us know any problems with the translator here in the forum.
  • zurpher
    The tag import works great now. =)

    I just noted a small issue that I wanted to report here:

    When importing the following entry:
    http://www.fachportal-paedagogik.de/fis_bildung/suche/fis_set.html?FId=1060389

    The keyword "Kounin, J." has been separated into two tags during import: "Kounin" and "J.". Quite frankly, this is not a big issue for me. However, I thought that only semi-colons ";" are used to split tags but it seems that commas "," are used too. I leave it up to you to decide whether this needs fixing.
  • zurpher
    Hello,

    I wonder whether the import from peDOCS can be improved.

    1) The attached PDFs are not imported automatically.

    2) The abstract is sometimes mixed up. For example with symbols and strings such as _&#8208_ and _‘_
    Ziel der diskursanalytischen Betrachtung von politischen Dokumenten (hierzu gehören: Pläne, Programme und Berichte in bildungs‐ und kindheitsrelevanten Politikfeldern) und des darin auftretenden Diskurses zum Thema gute Kindheit und Bildung ist es, sie daraufhin zu untersuchen, welche Vorstellungen/Wissensbestände von ‚guter Kindheit‘ (und damit verbunden richtiger Erziehung und Bildung, guter Elternschaft, guter pädagogischer Arbeit in Bildungsinstitutionen) transportiert werden. Das Projekt beabsichtigt daher, die politisch hervorgebrachten Leitbilder einer frühen und öffentlich verantworteten ‚Bildungskindheit‘ (vgl. Betz 2010) und ihre Konnotationen zu rekonstruieren. Ein Ausgangspunkt ist, dass die in den Dokumenten erzeugten Annahmen, Strategien und Maßnahmen einen weitreichenden Einfluss auf bildungspolitische Veränderungen und die Ausgestaltung der öffentlich verantworteten (Früh‐) Erziehung/Bildung haben, also soziale Prozesse aktiv mitgestalten, wie dies beispielhaft an der Implementierung der Bildungs‐ und Erziehungspläne oder der Initiierung von Elternbildungsprogrammen veranschaulicht werden kann (siehe auch Kapitel 4.3.2)). Weiter wird davon ausgegangen, dass diese wirkmächtigen Annahmen auch Einfluss auf die Strukturierung der Vorstellungen, Haltungen und Praktiken der sozialen Akteure haben, was es empirisch einzufangen gilt. Der Frage, ob und in welcher Art und Weise dies der Fall ist, inwieweit hier die sozial situierten im Sinne von milieuspezifisch unterschiedlich geprägten Vorstellungen ‚guter Kindheit‘ mit den Vorstellungen in politischen Berichten korrespondieren und welche Bedeutung dies wiederum für die Reproduktion von Bildungsungleichheiten hat, soll [hier] explorativ nachgegangen werden. (DIPF/Orig.)
    Instead of:
    Ziel der diskursanalytischen Betrachtung von politischen Dokumenten (hierzu gehören: Pläne, Programme und Berichte in bildungs‐ und kindheitsrelevanten Politikfeldern) und des darin auftretenden Diskurses zum Thema gute Kindheit und Bildung ist es, sie daraufhin zu untersuchen, welche Vorstellungen/Wissensbestände von ‚guter Kindheit‘ (und damit verbunden richtiger Erziehung und Bildung, guter Elternschaft, guter pädagogischer Arbeit in Bildungsinstitutionen) transportiert werden. Das Projekt beabsichtigt daher, die politisch hervorgebrachten Leitbilder einer frühen und öffentlich verantworteten ‚Bildungskindheit‘ (vgl. Betz 2010) und ihre Konnotationen zu rekonstruieren. Ein Ausgangspunkt ist, dass die in den Dokumenten erzeugten Annahmen, Strategien und Maßnahmen einen weitreichenden Einfluss auf bildungspolitische Veränderungen und die Ausgestaltung der öffentlich verantworteten (Früh‐) Erziehung/Bildung haben, also soziale Prozesse aktiv mitgestalten, wie dies beispielhaft an der Implementierung der Bildungs‐ und Erziehungspläne oder der Initiierung von Elternbildungsprogrammen veranschaulicht werden kann (siehe auch Kapitel 4.3.2)). Weiter wird davon ausgegangen, dass diese wirkmächtigen Annahmen auch Einfluss auf die Strukturierung der Vorstellungen, Haltungen und Praktiken der sozialen Akteure haben, was es empirisch einzufangen gilt. Der Frage, ob und in welcher Art und Weise dies der Fall ist, inwieweit hier die sozial situierten im Sinne von milieuspezifisch unterschiedlich geprägten Vorstellungen ‚guter Kindheit‘ mit den Vorstellungen in politischen Berichten korrespondieren und welche Bedeutung dies wiederum für die Reproduktion von Bildungsungleichheiten hat, soll [hier] explorativ nachgegangen werden. (DIPF/Orig.)
    Oh well, it does not show up. Perhaps because it is HTML code not imported correctly into Zotero.

    See: http://nbn-resolving.de/urn:nbn:de:0111-pedocs-118379

    3) I also noticed that the import sometimes does not work with some entries but have not appropriate example at hand at the moment.
  • noksagt
    It looks like the regex for the fachportal translator should be expanded to include pedocs, but this still won't fix (1).

    It would seemingly fix (2), as the problem seems to happen due to the COinS translator being used on pedocs.de. See:
    http://www.fachportal-paedagogik.de/fis_bildung/suche/fis_set.html?FId=A19580&mstn=1&ckd=no&mtz=200&facets=y&maxg=5&suche=einfach&ohneSynonyme=y&sort=jahrAb&feldname1=Freitext&feldinhalt1=+DISKURSANALYTISCHE+REKONSTRUKTION+POLITISCHEN+LEITBILDERN+BILDUNGSBEZOGENER+'GUTER+KINDHEIT'&bool1=and&next=950985&prev=&nHits=2&marker=1

    COinS should be UTF-8 url-encoded. I haven't checked, but I imagine that a different character set is being used on that page.

    Can't help with (3) without examples.
  • zuphilip
    Yeah, I think the problem/difference is that for the Fachportal Pädagogik we can use the BibTeX, but this option does not exists for pedocs. There is only the possibility to export in Endnote format (enw), which is AFAIK not supported by Zotero. The COinS format which is then used instead, seems to miss some things...
  • adamsmith
    enw is what Zotero calls Refer/Bibix
  • zuphilip
    Ah, I see. It should be possible to use this format then. However, it looks to me, that this would be a new translator for pedocs because the two things are too different. I will give it a try...
  • zuphilip
    A translator update is available now. This should handle pedocs as well (pdfs and multiple), fix encoding and drop the snapshots.

    Your version of Zotero will automatically update within 24hs, or you can update manually using the "Update Now" button in the "General" tab of the Zotero preferences. If you're using Standalone, restart Zotero and your browser after updating. Any further problems let us know & thanks for reporting.
  • zurpher
    Thank you, zuphilip!

    I just noted a small issue that I wanted to report here:

    When importing the following entry:
    http://www.fachportal-paedagogik.de/fis_bildung/suche/fis_set.html?FId=1060389

    The keyword "Kounin, J." has been separated into two tags during import: "Kounin" and "J.". Quite frankly, this is not a big issue for me. However, I thought that only semi-colons ";" are used to split tags but it seems that commas "," are used too. I leave it up to you to decide whether this needs fixing.
    I think the above issue is not fixed yet. I just imported the following entry and all tags with a comma in between have been separated into two tags.

    http://www.fachportal-paedagogik.de/fis_bildung/suche/fis_set.html?FId=991197

    In this case, these were quite a few. See:
    Schlagwörter: Pädagogik; Reformpädagogik; Progressive Erziehung; Bildungsgeschichte; Schulreform; Schulgeschichte; Schulentwicklung; Laborschule; Lehrplan; Unterrichtsgestaltung; Unterrichtsmethode; Unterrichtsorganisation; Unterrichtsreform; Geschichte (Histor); 19. Jahrhundert; 20. Jahrhundert; Adler, Felix; Dewey, John; Parker, Francis W.; Parkhurst, Helen; Washburne, Carleton; Chicago, Ill.; New York, N.Y.; USA
  • zurpher
    Dear zuphilip,

    when importing from peDOCS, is it possible to automatically add the URN as URL in Zotero?

    See for example the following entry:

    http://www.pedocs.de/frontdoor.php?source_opus=3437

    It comes with the associated URN:

    http://nbn-resolving.de/urn:nbn:de:0111-opus-34378

    Currently, the URN is not imported. It would be great if this could be added.

    Many thanks for your great support!
  • zurpher
    The German Education Index "FIS Bildung" has been relaunched. Unfortunately, with the relaunch the item import does not work anymore. Can this be fixed please?
  • zurpher
    I just wanted to add that pedocs still works. See for example: http://www.pedocs.de/frontdoor.php?source_opus=4468

    Maybe this helps with the trouble shooting.
  • zuphilip
    I have seen the announcement for the redesign: http://www.inetbib.de/listenarchiv/msg61735.html and created a ticket for Zotero translator.

    Today, I see that COinS is working. Can you confirm that? I haven't looked closer, how good the data is at the moment.
  • zurpher
    The item import is possible again but the "Schlagwörter" are not imported as Tags. It would be great if this could be restored too.

    See for example:
    http://www.fachportal-paedagogik.de/literatur/vollanzeige.html?FId=1041537#vollanzeige
Sign In or Register to comment.