get Meta-Data from PDF doesn't work for Thieme DMW

cb102
edited January 30, 2017
As an example
https://www.thieme-connect.com/products/ejournals/html/10.1055/s-2007-959047

Zotero is able to add this to its library without problems.

But when I just have the PDF-File (maybe from my university) and try to do "get Meta-Data from PDF" it doesn't work. Not sure which part of Zotero is responsible for parsing PDF files.

There is the same problem with this paper from JAMA
http://jamanetwork.com/journals/jama/fullarticle/1167317
  • adamsmith
    Not much to do here -- getting metadata for PDF involves a bunch of sophisticated guesswork, so it's less reliable than getting data from the journal's website or a database. The two are unrelated.
  • cb102
    I agree with you that it is "less reliable". But sometimes you need it to do it that way. This isn't an argument not to take care of it here. If you think that feauter doesn't make sense you should kill it from Z.

    What do you mean the "two are unrelated"?. The journals are not important enough? You looked at the IF? Or you are a part of the scientific community related to that two journals? They are damn related here.

    I know about your less resources. Because of that you can say: We don't take care of it. We have other priorities.

    But your arguments IMO to overbearing (not sure if that is the correct word). It is an attitude I wouldn't like to find in an open source project - especially if it is such a damn important one as Z is.
  • adamsmith
    geez -- not everything is a big philosophical debate about the nature of open source. Also, no need to swear.

    What I'm saying is that retrieving metadata for PDFs relies on algorithms that perform a good guess as to what the metadata is. They work a lot of the time, but sometimes they don't. Whether they work or not has nothing to do with whether Zotero is able to import from the journal's website (i.e. "unrelated"), since Zotero queries central databases for PDFs, mostly CrossRef and Google Scholar.

    Because of the nature of these queries, where they don't work, there's currently not much we can do. I assume you agree that this is still a useful feature even if it only works about 80% of the time.

    All that said, those two articles actually do work for me. What's the exact error message you're getting? What happens when you paste 10.1055/s-2007-959047 into the add by identifier window?
Sign In or Register to comment.