Wikipedia...

I'm sorry if this feature has already been requested, but I did search it in the forums, and the only result that came up was not exactly the idea I had in mind.

My request is for the Zotero icon to be available for Wikipedia articles. I know I used Wikipedia as a resource for several papers I wrote in college, and my professors widely accepted its use in papers. That was several years ago, so I'm sure that now, Wikipedia is probably even more widely accepted as a source for college research.

In any case, I think it would be very helpful for Zotero to be able to generate an entry from a Wikipedia article page. It seems that the information could be easily obtained, as all Wiki articles are in a standard format, with the critical information located in the same place every time (or atleast that is how I understand Wikipedia to be organized).

I understand that it would be difficult for some of the information that Zotero usually lists to be pulled from a Wikipedia page. However, some basic information could be gathered, such as Title, Address, Date Accessed, ect.

I understand that any Zotero user can create an entry with general ease, but if other users use Wikipedia as much as I do (and did in college), it would be very convenient to have such information already accessible by using the Zotero icon in the address bar.

Thank for reading my request. I would love to help on the development of such a feature, but I know nothing about making such tools, so I probably would not be very useful.

-MK
  • There has been some talk aout embedding metadata on WP pages to make this easier.

    "Title, Address, Date Accessed, ect." are added when you click the "Create New Item from Current Page" button. Were you aware of this feature?

    If you were, it'd be useful to refine your request. If the "create new item from current page" is satisfactory, but you find using the Zotero icon in the URL button is "easier," perhaps you want an option to put an icon up for EVERY page (preferably one which was different from the icon that Zotero has site translators for).

    If it isn't satisfactory, how should a Wikipedia-specific translator be different? A few ideas:
    • rm " - Wikipedia, the free encyclopedia" from title
    • Place "Wikipedia in "Publication"
    • Use the URL corresponding to the current revision.
  • I added COinS metadata to the "Cite this article" page, so you should be able to get the very basics of the article into Zotero. See this example.
  • Can you please be more specific as to the usage. Are you saying we can push an article into Zotero with Cite this article? Or are you saying we that when I Create New Item From Current Page I will have more fields properly populated?

    Either way, I think I am missing something, perhaps a lot, and look forward to a clarification.
  • @dme: Who was your query directed to? mousekitty did not state exactly what kind of citation was wanted. I pointed out that Zotero is already able to cite ANY URL as a webpage. If more citation info is needed for WP than is already provided, we should discuss what kind of information so that either WP can start embedding it or Zotero can start scraping it or both.

    Omegatron added embedded data to the citation templates on WP. Thus, if you go to a page which has citations (like <http://en.wikipedia.org/wiki/Atom_probe>;), you'll be able to grab those references.

    Finally, the new version of Zotero has a useful quick-copy feature, that allows you to copy a reference from Zotero & paste it in the WP citation template format onto WP.
  • Omegatron added embedded data to the citation templates on WP. Thus, if you go to a page which has citations (like ), you'll be able to grab those references.
    Yes, but that's not what I was talking about. Mousekitty asked for the article itself to be entered into Zotero. I added COinS tag for the article itself, too, but you have to click the "Cite this article" link in the toolbar, as shown in my example link. The COinS is at the very bottom.
  • Perhaps it would make sense to embed the article citation COinS tag on the main page for all articles? Any included references would also show up as importable citations.
  • Perhaps it would make sense to embed the article citation COinS tag on the main page for all articles? Any included references would also show up as importable citations.
    It definitely would, but that would require a software change.
  • Ok, ok. I see that I am swimming, I mean treading water in the deep end.

    Now we all know I am a newbie.

    1. I was confused by the lack of any reference or any button on the Special:Cite page ... I now realize why this is the case, but I can also see how others might be similarly confused. Yes, yes I know, I should tell Zotero to "Create New Item From Current Page" to grab the WP page into my collection. I think you're going to tell me that a Cite in Zotero button on the Special:Cite page is not going to fly, but how about including in the Important Note something about Zotero and similar tools and one line on how to add the WP page to your collection.

    I love that work that Omegatron did (am I correct if I say, "Omegatron has embedded data in the citation templates on WP so that WP pages now display a nifty icon which when clicked will list all sources referenced in any WP article"?

    What is the proper term for this "nifty icon"?

    2. When I "Create New Item From Current Page" from a WP page Zotero creates a link to the current page. I'd like to set Zotero to create a Permanent link to the current page. That is, if the page is a Permanent page, then link to that page, if the page is the article, then link to the Permanent version of that page. Yes, I could do this myself, by clicking on Permanent link, but I'd prefer it if Zotero had an option that I could enable to perform this.

    3. Is there any way, or any structured way, to insert WP content from a Discussion, I mean Talk, page into Zotero?

    4. I have not yet thought about, much less used Zotero to insert content in to the WikiPedia, so I'll discuss that at a later date.

    4a Is there any need to push data from Z to a Talk: page? I imagine there is, so is there a structured way to do so?

    5. I think mousekitty wants a robust way to grab data from WP into Z. Sure it would be nice if Z populated the Website Title field and other fields in the Web Page Template, but I think mousekitty was asking for a WP specific Template which includes, say, a reference, or flag, for any images; # of revisions, talk page flag, list of contributors, # of revisions, creation date, date of last contribution, etc. If this is possible or well received I'd be happy to contribute thoughts, specs, etc., but I don't code.

    NB. Ambiguity Alert: WP=Wikipedia and WP=WordPress
  • edited April 3, 2007
    I was confused by the lack of any reference or any button on the Special:Cite page ... I now realize why this is the case, but I can also see how others might be similarly confused.
    I do not understand why this is the case. Can you explain in more detail?
    • If you view a regular Wikipedia article without citations, Zotero will not detect anything. There should be no icon in the address bar.
    • If you view a Wikipedia article with certain types of citations (in a References section, most likely), Zotero will pop up an icon in your address bar that lets you download the citation(s) into Zotero. In other words, if you view the article about Abraham Lincoln, you can download info about various Abraham Lincoln books into your Zotero.
    • If you view a Wikipedia article and then click the "Cite this article" link in the Wikipedia navigation bar, you will go to a special page that has bibliographic information about that particular revision of the Wikipedia article itself. You should see a "journal" icon in the address bar that will let you download info about the Wikipedia article itself, even though it's not a journal (websites and similar resources will supposedly be added to COinS in the future).
    Yes, yes I know, I should tell Zotero to "Create New Item From Current Page" to grab the WP page into my collection.
    That will just save it as a generic website with no author information or anything. If Zotero knew how to scrape Wikipedia articles, it could deduce all the information that is normally displayed on the "Cite this article" page and save it as an encyclopedia article.
    Yes, I could do this myself, by clicking on Permanent link, but I'd prefer it if Zotero had an option that I could enable to perform this.
    I think mousekitty wants a robust way to grab data from WP into Z. Sure it would be nice if Z populated the Website Title field and other fields in the Web Page Template, but I think mousekitty was asking for a WP specific Template which includes, say, a reference, or flag, for any images; # of revisions, talk page flag, list of contributors, # of revisions, creation date, date of last contribution, etc.
    Someone could write a Zotero scraper for it, or we could add COinS to the article itself by asking the developers to modify Mediawiki. Since the "journal" format isn't really appropriate, though, I'd want to wait until the correct format is settled on before changing the software. I guess we could ask for a Mediawiki: namespace page and fill in the info with magic words so that admins could modify it in the future...
    I think you're going to tell me that a Cite in Zotero button on the Special:Cite page is not going to fly, but how about including in the Important Note something about Zotero and similar tools and one line on how to add the WP page to your collection.
    Where?
    am I correct if I say, "Omegatron has embedded data in the citation templates on WP so that WP pages now display a nifty icon which when clicked will list all sources referenced in any WP article"?
    It will display a nifty folder icon in the address bar that lets you download the information about all citations in the article (but not the article itself).
    I have not yet thought about, much less used Zotero to insert content in to the WikiPedia, so I'll discuss that at a later date.
    It will be possible! You can export citation templates with the latest beta.
  • edited April 3, 2007
    or we could add COinS to the article itself by asking the developers to modify Mediawiki. Since the "journal" format isn't really appropriate, though, I'd want to wait until the correct format is settled on before changing the software.
    I think embedding self-describing COinS on a web page is a dirty hack (mostly because of the relative narrowness of the OpenURL spec, such as the typing that you bring up). I think it'd be great to get changes into MediaWiki, but would have a slight preference for embedded RDF and/or unAPI. I think there's already been work on RDF extensions, but haven't followed any of it closely.
  • I think embedding self-describing COinS on a web page is a dirty hack (mostly because of the relative narrowness of the OpenURL spec, such as the typing that you bring up).
    Ok. Is it appropriate for the "Cite this article" page? I was told it was on the gcs-pcs-list.
    I think it'd be great to get changes into MediaWiki, but would have a slight preference for embedded RDF and/or unAPI. I think there's already been work on RDF extensions, but haven't followed any of it closely.
    I don't know much about those.
  • Perhaps we can move this piece of discussion onto WP somewhere, so we don't hijack Zotero's forums?
    Ok. Is it appropriate for the "Cite this article" page?
    For lack of a better alternative, I suppose it is.

    Most pages which have embeddded RDF use meta tags in the page header--this paradigm is obviously used to write metadata about the current webpage, and might be useful for clients other than Zotero (which self-describing COinS entries don't really seem to be). While RDF in HTML can also describe other things, I don't know if there is a strong case to use it over COinS on the "Cite this article" page.

    unAPI might be cool to have on the "Cite this article" page, as it could theoretically just reuse the BibTeX & so any enhancements to "Zotero support" would benefit bibtex users & vice versa.
  • Whew...things are cooking in this discussion, though it may have moved to the WP. If it has can you please tell me where it has gone.

    Journal Icon: Looks like a document, but populates the Journal template. Ok. Now, what's the name for the folder icon that generates multiple entries based on References?

    Confusion, is unfortunately a state that many people find themselves in quite a lot. As for what's appropriate where...just look at the top of this page where you'll find a button entitled, "Bookmark this dicsussion". If I click that button will it add a bookmark to my Browser's Bookmarks? Will it add a reference to Zotero? Turns out it doesn't do either.

    So it's hardly surprising that I was confused about the WP: Special:Cite. I am less confused now, however, and understand that that page is designed for someone to copy by hand, but now that Omegatron has written some code and added it to the WP I owe a debt of gratitute for making my life and those of other WP and Zotero users a little easier. Furthermore, I think I understand what you are talking about re: WP and Journal. So who do we have to contact re: a new format for MLA, et al.?

    Now if only someone could write a plain English description of COinS I might be able to progress still further.

    This is not English:
    ContextObjects in Spans, commonly abbreviated COinS, is a method of embedding latent OpenURL ContextObjects in web pages. This allows client software to retrieve bibliographic metadata and to use an OpenURL resolver to find a mediated link. A principle advantage of using COinS, rather than giving a static OpenURL, is that the client can determine which resolver to use. This allows, for instance, searching for a copy of a book in one's own library.

    Do you have any clue as to the time problem described here: http://forums.zotero.org/discussion/669/middle-column-date-addedis-wrong/#Item_1

    Hoping to learn still more from the Gods.
  • Humm....So are you saying that if there was a Wikipedia category in the MLA et al. We could use Locate? That is to say that, "Locate" wouldn't try to find the item in a library catalog, but would know to use the url to Locate the item?
  • edited April 4, 2007
    Most pages which have embeddded RDF use meta tags in the page header--this paradigm is obviously used to write metadata about the current webpage, and might be useful for clients other than Zotero (which self-describing COinS entries don't really seem to be). While RDF in HTML can also describe other things, I don't know if there is a strong case to use it over COinS on the "Cite this article" page.
    It seems there are several overlapping concepts here. Since anything other than COinS can't be done on a user level, we need to decide exactly what we want and ask the developers. Wikipedia:WikiProject Microformats is related, and maybe the right place to discuss this further?
    This is not English:
    ContextObjects in Spans, commonly abbreviated COinS, is a method of embedding latent OpenURL ContextObjects in web pages. This allows client software to retrieve bibliographic metadata and to use an OpenURL resolver to find a mediated link.
    Agreed. :-) I'm the major person adding these things to Wikipedia, and I don't even really know what they are; I just know what they're useful for.

    In my current understanding, a COinS tag is a little snippet of hidden, but computer-readable, bibliographic information, that can be used in a variety of ways. It uses the same format as the last part of an OpenURL, but encodes it so that it looks like plain text to a browser. OpenURLs are a way to link directly to something like a book or journal article, including a first part (which is specific to your school's library) and a second part (that describes the book with things like author and title). Your school's library "resolver" receives that information when you follow the link, figures out which book it is, and directs you to the page that gives information about that book.



    Edit:
    Now that I look at it, we can do microformats from userspace, as demonstrated on http://en.wikipedia.org/wiki/HCard#Live_example , since it's just divs with special classes. The citation one isn't finished yet, though?
  • Since anything other than COinS can't be done on a user level,
    ANY self-describing format (COinS included) can't be done on a user level.

    I see no reason why embedded RDF which wasn't self-describing couldn't be added to the citation page (though I don't know how well Zotero imports this right now--it didn't the last time I checked). I don't know how this page was created--unAPI support on that page might not be inconceivable.
    Wikipedia:WikiProject Microformats is related, and maybe the right place to discuss this further?
    Sounds good to me--kick it off & provide a link.
    Agreed. :-) I'm the major person adding these things to Wikipedia, and I don't even really know what they are; I just know what they're useful for.
    As the originator of that page, I feel some sense of responsibility for any confusion. I'm reasonably familiar with the OpenURL spec. However, it is a technical topic & esoteric enough that relying on the OpenURL article, etc. to bring people up to speed seems unavoidable. Please raise any questions on the talk page for the article & we'll try to address them.
    Now that I look at it, we can do microformats from userspace, as demonstrated on http://en.wikipedia.org/wiki/HCard#Live_example , since it's just divs with special classes. The citation one isn't finished yet, though?
    Yes, hCite is still being discussed. As I said, we can also already embed RDF in userspace.

    We can't do anything which is self-describing and automatic without WP developers buying into it.

    If our goal is merely to get data into Zotero, a home-grown site translator seems like the easiest method. If we want to make WP data useful to other tools too, unAPI holds a lot of promise & embedded RDF is marginally more "mature" and might be "good enough."
  • ANY self-describing format (COinS included) can't be done on a user level.
    Well, they could be, but they'd have to be done manually on each page, which would be dumb.

    But the citations can be done in userspace, as long as the code is compatible with wikimarkup. (COinS are, microformats with div are, but not ones with abbr)
    Sounds good to me--kick it off & provide a link.
    I meant to link it.

    http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Microformats
    we can also already embed RDF in userspace.
    I need to read more about it, I guess. It looks like XML namespace stuff, which wouldn't be accepted by the software.
  • embedded RDF doesn't use "XML namespace stuff." It uses (X)HTML tags, just like any other microformat. This can be entirely in SPAN elements (a'la COinS) if desired.
  • Clearly the Gods are speaking, albeit in tongues, but some of the bits are in cleartext (Who knew the Gods had a sense of humor):
    Agreed. :-) I'm the major person adding these things to Wikipedia, and I don't even really know what they are; I just know how to use 'em
    I'm beginning to understand COinS, but OpenURL...well that's not quite understandable:
    OpenURL is a type of URL that contains resource metadata for use primarily in libraries. The National Information Standards Organization (NISO), has developed OpenURL and its data container (the ContextObject) as international ANSI standard Z39.88. On 22 June 2006, OCLC was named the maintenance agency for the standard.
    But I'm working on a definition...see: http://en.wikipedia.org/wiki/Talk:OpenURL
    How can a mere mortal install this in their own MediaWiki?
  • I like how y'all told the world that we have COinS for all our references now, when, uh... we don't yet.

    Oh well. Time to get our butts in gear.
  • Some MediaWiki sites (including Wikipedia) do have an OAI-MPH repository interface. Perhaps this could be used to retrieve citation metadata (as it is for arXiv, etc.)? The caveat is that the interface isn't public--it requires authentication. But perhaps someone on the Zotero team can talk to Brion Vibber about getting an account to test the interface?
Sign In or Register to comment.