Citation metadata not pulling into Zotero correctly for our site

I am the product manager for AccessEngineering at McGraw-Hill Education. We are working on a relaunch of our content site. We have tagged our content with google scholar tags so that they will hopefully import the metadata correctly for users that use the Zotero browser plug-ins. But we are finding that our chapters are not coming into Zotero correctly and we are not sure why. Specifically, it looks like both books and chapters are being identified as "books" so the chapters are not capturing both the book name and the chapter name separately. Because we are currently on a closed development site, we've taken screenshots of the source code and the content page and the Zotero output to show you and loaded them here: https://drive.google.com/drive/folders/16tq-8UU2btuip3o4sMNtbnIaNya-SLaT?usp=sharing

I'm hoping you can tell us how to get this corrected.

Thanks,
Lauren Sapira
lauren.sapira@mheducation.com
  • Zotero is looking for citation_book_title not citation_inbook_title in the metatags. I'm not aware of any place with a comprehensive and authoritative list (https://scholar.google.com/intl/en/scholar/inclusion.html#indexing includes neither) and if you can point to a place that suggests inbook is reasonably common, I don't see any reason for us not to include it -- if you just picked it lacking a better alternative, I'd suggest going with citation_book_title.
  • Thanks - so if they change this, will Zotero be able to pick up both the chapter title and the book title with the way we have the other tags? We want to make sure Zotero can pick both of these up correctly.
  • Hi, are you able to respond on my follow-up question? I want to give this feedback to the platform vendor but just wanted to know if it seems like we have the rest set up correctly for chapter level citations to pull in correctly.
  • I'm fairly sure, yes -- a bit hard to do this without a site to check on though.

    Obviously the publisher tag is empty and it's tricky to get editors imported via generic metadata, but getting this recognized as a book chapter with the title and book title recognized correctly should work with that change.
  • Ok thanks - yes, I saw the publisher tag as empty and already have them fixing that.
  • Hi - they say they made the change you suggested above but the citations are still not coming in correctly into Zotero with the chrome plugin. You can get to our test site here - could you see if you can tell what they are doing wrong? They think they have the tags set up correctly but clearly they do not.

    http://mhe.stage.highwire.org
    Username: mhe
    Password: !!mhe19!!

    some url's you can try:

    http://mhe.stage.highwire.org/content/book/9780071830829/chapter/chapter2
    This is a chapter of a book. When I use the plugin, it seems to be pulling in data though from a video and not from the chapter page?

    http://mhe.stage.highwire.org/content/video/V2939922913001
    this is a video, but it says it's a webpage and not all of the metadata is pulling in correctly either.

    http://mhe.stage.highwire.org/content/book/9781259643835/toc-chapter/chapter1/section/section3
    This is a section of a book. It is pulling in the section name as the book title and it doesn't pull in the actual book title or the author.

    http://mhe.stage.highwire.org/content/book/9780070704442/chapter/chapter2
    This is a chapter of a book. Again the chapter title is being pulled in as the book title and it's missing the book title and the author.

    thank you for your help.
  • Hi Adam - are you able to take a look at this for us? Thank you so much for your help.

    Lauren
  • Don't have time to look at all of these, but e.g. for the first link, here's what they have in the metaheader:
    <meta name="citation_publication_date" content="2019-02-21" />
    <meta name="citation_publisher" content="McGraw-Hill Education: New York, Chicago, San Francisco, Athens, London, Madrid, Mexico City, Milan, New Delhi, Singapore, Sydney, Toronto" />
    <meta name="citation_public_url" content="/content/book/9780071830829/chapter/chapter2" />
    <meta name="citation_isbn" content="9780071830829" />
    <meta name="citation_book_title" content="Schaum&#039;s Outline of Thermodynamics for Engineers, Third Edition" />
    <meta name="citation_title" content="Properties of Pure Substances" />
    <meta name="citation_author" content="Thom Adams, Ph.D." />
    <meta name="citation_author" content="Thom Adams, Ph.D." />
    <meta name="citation_abstract" content="&lt;para&gt;This video demonstrates the use of the steam tables to find quality as well other thermodynamic properties.&lt;/para&gt;" />
    <meta name="citation_abstract" content="&lt;para&gt;This video demonstrates the use of various equations of state and the use of the generalized compressibility chart.&lt;/para&gt;" />
    <meta name="DC.Title" content="Properties of Pure Substances" />
    <meta name="DC.Created" content="2019-02-21" />
    <meta name="DC.Format" content="text/html" />
    <meta name="DC.Identifier" content="urn:isbn:9780071830829" />
    <meta name="DC.Type" content="text" />
    <meta name="DC.Description" content="This video demonstrates the use of the steam tables to find quality as well other thermodynamic properties." />
    <meta name="DC.Description" content="This video demonstrates the use of various equations of state and the use of the generalized compressibility chart." />
    <meta name="DC.Publisher" content="McGraw-Hill Education: New York, Chicago, San Francisco, Athens, London, Madrid, Mexico City, Milan, New Delhi, Singapore, Sydney, Toronto" />


    So you see that the video issue is a problem in what they put on the page. I'm not 100% sure what' the problem with the book vs. chapter issue is, but the password protection also makes this very cumbersome to test, so I'm afraid I'm not going to be able to spend more time on it.
  • edited March 15, 2019
    I, too, tried to look at your page headers because I am always trying to improve the metadata available from my online index (apart from Zotero). I echo adamsmith's comment about passwords. Each time I tried to view a page or to view the source code to see the header I had to renter the password.

    I do have a few recommendations. When you determine what to put in your DC or GS/Highwire tags think of how those tags will be used. Probably there are 2 main purposes: 1) indexing by Google Scholar and 2) so that your metadata can be properly imported into bibliography management software (like Zotero) and properly cited when future authors reference your material.

    In the example (above) there is much unnecessary content and unusable content. The publisher tag contains the publisher name and muliple cities as the publisher location. I know of no citation style that allows 12 publisher locations. This means that the software will need to both parse the location from the publisher name and that the software user will need to edit the location to remove irrelevant cities.

    Your GS tags include both the book title and chapter title but the Dublin only includes the chapter title.

    Several tags are repeated. This isn't necessarily always a problem but it can be confusing when the author tag is repeated: Is the listed author responsible for both the chapter and the entire book? Are there two generations of authors with the same name? Each of the two tag flavors contains two similar but different abstract or description tags. [Aside: You should choose whether you want to use a brief marketing sentence or a more scholarly summary of the chapter.]

    I am not alone in objecting to including academic credentials with the authors' names. Also inappropriate are things like 'professor', 'chaplain ', 'doctor', etc. Those seem to often end up being included among the author's initials.

    Someone with more expertise than I have will need to comment about this final recommendation but I feel that it is usually best to keep separate bibliographic indexing metadata from search engine Google Scholar tags that are useful to locate items online. For indexing I suggest using a system such as UnAPI. Another option would be to link to a tagged format (such as RIS) where you can break-out each metadata item. If your pages are consistent and the RIS link position doesn't require intermediate steps, Zotero can find and use that file for importing your metadata.
  • On the last point -- yes and no. It's absolutely the case that unAPI or related functionality allows you to provide richer metadata (Zotero is working on allowing for more common/modern approches to do the same such as JSON-ld and link rel="altenate" type=""), but you should also be able to get the basics of standard item types into Zotero using GS Highwire tags.
Sign In or Register to comment.