Drag and Drop...and extract references from a scientists articles

grupoUIS
If Zotero succeed in extract the references of an article that would be great, everyone would switch to Zotero this feature has never been seen before. Another thing that would be great is to be achieved to further integrate with office, so there may be a floating bar on the side of Microsoft Word and we could simply drag a reference to the text and display the reference number (eg " this is a test document (1)", and then the complete reference appear below)... something like that would save us so much time ..

Thanks, sorry for my English, but now I'm using google translator (I have lazy)...i hope all this changes appear soon in the big Zotero...bye
  • ivotron
    I second this. Automatically extracting references for an article would be pure awesomeness. This may not be useful for some people, in which case a dialog can be added where Zotero asks whether or not to extract references.
  • mronkko
    Extracting references from an article reliably is quite difficult. Although Zotero has the ability to scrape the information from websites, doing this for articles would be difficult. Also in my case I want to have the PDFs too, so I would need to go to get the articles from a database anyway.

    Moreover, this could be done as an add-on to Zotero, so I think that it is unlikely that the developers would implement this.
  • ajlyon
    Try out http://freecite.library.brown.edu/welcome . Copy and paste the bibliography section from the PDF you are using and that site will try to parse it. Zotero can import those references that can be parsed.
  • adamsmith
    if it's mainly articles there is a similar service on the crossref site - freecite has some problems still, particularly with works published prior to 2000
  • rotofils
    You can also try http://refformer.com
    It works only with journal references.
  • ajlyon
    If you have contact information for the author of refformer.com, you may want to encourage them to add standard export formats, like BibTeX, RIS or COinS, so that Zotero and other tools work more effectively with it.
  • rotofils
    In RefFormer.com registered user can save a reference and export it easily to Zotero by clicking "Tagged"
  • adamsmith
    actually, the saved list of references has COinS already implemented. Works quite nicely - too bad it only does journal references, but that seems to work pretty well - cool!
  • ajlyon
    That's good to hear -- I had tested it without creating an account. Still, there's no reason not to embed COinS for guest users as well.
  • rotofils
    I do not provide any more information for anonymous user because I'm afraid of robots exploiting it.
  • ajlyon
    Do you mean that you are concerned that someone will try to use your software to do automated citation parsing in an automated manner? Since you provide the parsed data in a _reliable_ citation format (Chicago, IEEE, etc), that is functionally equivalent to providing COinS, as far as a potential robot citation parser is concerned. It is not equivalent for browsers that can take advantage of embedded markup, like Firefox+Zotero and hopefully some other systems in the future.

    It's clearly your right to run your site as you wish, but I think that you're losing a good chance to be a good citizen of the semantic web in return for a false sense of protection against exploitative usage.
  • rotofils
    Well, thank you for the preaching, but I think we are going off topic.
  • dstillman
    rotofils: If you're going to come and spam the forums with a link to your own site without initially disclosing your connection to it, you should probably be prepared for people to comment on it rather than making snide remarks when they do.
  • rotofils
    I'm sorry. I should not make the remark. I felt the part about "loosing a good chance to be a good citizen" not appropriate.
  • tesolchina
    I think one possible solution is to create a new file format for scientific literature where the authors would save the references in a way that the machine can read the information without parsing. Currently, we use zotero or other software to convert the bibliographic info in machine-readable form to plain text to be published in a journal. And then we try to convert the plain texts back to a format that the machine can read. What if we bypass the whole conversion thing and keep the original machine-readable format when we submit our manuscripts. Of course, that would require a file format different from PDF or DOC. Zotero seems to be in a good position to create such a new format.
    Simon
  • beogl
    @rotofils: I just had a look at your site. It looks useful, but I would like to use reference parsing locally on my pc.

    Before I found this discussion I started writing on some code for that, but maybe I can use yours. Is your code GPL or do you prefer to keep it private? (I will make mine GPL.)
  • beogl
    Seems like there already are a lot of free parsers. It looks like there are some useful references here:

    iSpiders: Reference Parser Revived
    http://ispiders.blogspot.com/2010/08/reference-parser-revived.html

    However I do not know much about the quality of them. Do anyone here know or have other tips?
Sign In or Register to comment.