a, b and c missing for no date

styrmirm
When you have a citation with same author same year you will have (Doe, 2011a) (Doe, 2011b) this is missing for no date, so there is no differenceation when there is the same author with no date.
  • fbennett
    edited March 31, 2011
    Nice one. I'll look at this today.
  • adamsmith
    agree that this needs a solution, but what exactly? I don't recall ever having seen
    (Smith n.d. a; Smith n.d. b) and it looks very weird to me.
  • fbennett
    Partially awake now (coffee still to come). Um ... yes, that would be a bit weird, wouldn't it. I guess that's where we start; what should the citation with descriminator look like in this case?
  • ajlyon
    And do any style guides even specify the correct behavior here?
  • styrmirm
    In the APA guidelines I have, written in Icelandic. They have a guide for this in a section about how to cite to the Internet, most likely because Internet is where this will probably occure the most.

    "When citing more than one documents with the same author that have no date, differanciate between the documents with letters (a, b, c) like done with other citations. For example ...on the homepage of Reykjavik city this is often mentioned (Reykjavikurborg, n.d.-a, n.d.-b)..."
    Translation is mine.

    Hope this help.
  • styrmirm
    I just noticed that there is in fact showed a descriminator between those citations in the biblography, but I am presuming that is an error most likely a bug. The descriminator is showed in "Retrived". Like this:
    Retrived March, 31 2011a from ...
    Retrived March, 31 2011b from ...
    This does not happen when there is a year with the author. So the code probably looks for the first year it finds and adds a descriminator. I actually find this funny.
  • fbennett
    Yep, that's certainly what's happening. It's not particularly useful, so we'll look into ways of implementing year-suffix discrimination on undated entries. By design, if the year-suffix renders earlier on the citation, it won't be rendered on the access date.
  • fbennett
    edited April 1, 2011
    Actually, it should be possible to get exactly the effect you're after by adding the year-suffix element explicitly. It's just an ordinary variable in CSL 1.0, so it can be rendered anywhere. This doesn't require any modification to CSL; the specification allows it.

    The CSL for such an effect might look like this (the block below is from a test fixture that I constructed this evening, to determine whether the processor can currently satisfy the requirement -- which it can't, without a small modification (see below): <citation disambiguate-add-year-suffix="true">
    <layout delimiter="; " prefix="(" suffix=")">
    <group delimiter=" ">
    <names variable="author">
    <name/>
    </names>
    <group>
    <choose>
    <if variable="issued">
    <date variable="issued" form="text"
    date-parts="year"/>
    <text variable="year-suffix"/>
    </if>
    <else>
    <text term="no date" form="short"/>
    <text variable="year-suffix" prefix="-"/>
    </else>
    </choose>
    </group>
    </group>
    <group delimiter=" " prefix=" [" suffix="]">
    <text term="accessed" suffix=": "
    text-case="capitalize-first"/>
    <date variable="accessed" form="text"
    date-parts="year-month-day"/>
    </group>
    </layout>
    </citation>    (Edited to correct syntax errors in my original post)
    In the current version of the processor, this doesn't work, because the year-suffix variable does not test positive for content when a year-suffix is available. I've made a small change to the code that fixes this little bug, and I'll push it to a release later this evening. You can expect to see the revised version in the next release of Zotero (2.1.3).

    Thanks for this interesting problem. :)
  • styrmirm
    You are welcome, glad to be able to better Zotero, I really like it.

    If I understand you correctly, if I add this to the APA.csl after release of version 2.1.3 then this should work.
  • fbennett
    Yep. You may need to adjust the formatting a bit to get the exact effect you need. The code above (just replaced with the code straight from the test fixture, which I've just finished) produces this citation string:

    (John Doe n.d.-a [Accessed: June 01, 1965]; John Doe n.d.-b [Accessed: June 01, 2065])
    Note that the accessed dates differ. That's in there because I discovered another little bug in the processor while building the test; when these accessed dates were included (which they almost never are in the citation form used for disambiguating, but never mind), the citations were never detected as ambiguous. So we squashed two bugs today.
  • adamsmith
    edited April 1, 2011
    styrmirm - feel free to add this to the current general apa style, too (I'd upload the changes back to the repository if you want) - you can find it here.
    https://github.com/citation-style-language/styles/blob/master/apa.csl
  • styrmirm
    adamsmith - Had to think about how to do this one. Not the best one in .csl files. But I hope I did this right.

    Added two lines to <macro name="issued-year">, was not sure if it should be added to <macro name="issued"> ,so I skipped that one. Now it is like this:

    <macro name="issued-year">
    <choose>
    <if variable="issued">
    <date variable="issued">
    <date-part name="year"/>
    </date>
    <text variable="year-suffix"/>
    </if>
    <else>
    <text term="no date" form="short"/>
    <text variable="year-suffix" prefix="-"/>
    </else>
    </choose>
    </macro>

    Here is the link: https://gist.github.com/6e9f3b4aabd1e0a97c5b
  • styrmirm
    Sorry, It didn't work. But now I fixed it and tested it with two authors with no date and one author with no date, also same author with same year. It is the same link, but now fixed.
    https://gist.github.com/6e9f3b4aabd1e0a97c5b
  • adamsmith
    OK, this is now implemented on the github repository.

This is an old discussion that has not been active in a long time. Instead of commenting here, you should start a new discussion. If you think the content of this discussion is still relevant, you can link to it from your new discussion.