Still more bugs with Harvard and multiple authors...

Hello everyone,

I've been finding quite a lot of strange bugs in Zotero with the Harvard styles that are close to leave me to despair... Or having to change to Endnote, since I have to finish my thesis soon... I'll give the examples below, using 3 multi-author references.

Example 1. I have opened a new blank word page. Zotero is setted for Harvard format 1. Put 3 multi-authors' references and the results are below. First observation: why do some appear "et al." and other don't? They should be all "et al" in the short reference group and complete in the bibliography. Second observation: in the bibliography, actually two articles have the full authors' list but one has decided to appear as "et al"...
Third observation: you have noticed that I have put twice the reference to Milinski et al and they both look the same. This is about to change in my next example.

(Milinski et al. 2006)

(Milinski et al. 2006)

(Delbecq, Van den Ven & Gustafson 1975)

(Arentsen, Bressers & O'Toole 2000)


Arentsen, M.J., Bressers, H.T.A. & O'Toole, J., 2000. Institutional and Policy Responses to Uncertainty in Environmental Policy: A Comparison of Dutch and U.S. Styles. Policy Studies Journal, 28(3), p597, 15p.

Delbecq, A.L., Van den Ven, A.H. & Gustafson, D.H., 1975. Group Techniques for Planning. A Guide to Nominal Group and Delphi Processes., Glenview, Illinois.: Scott, Foresman and Company.

Milinski, M. et al., 2006. Stabilizing the Earth's climate is not a losing game: Supporting evidence from public goods experiments. Proceedings of the National Academy of Sciences, 103(11), 3994-3998.

-----------------------------------------------------------------------------------

Example 2. I simply went to "Zotero Set Doc Prefs" button and changed it from Harvard format 1 to Harvard format 2. First observation: now Milinsky&Cia shows up once with "et al" and the second time with all authors listed? What the heck...?
Second observation: in the bibliography, all authors' were condensed into the "et al" format, when they should have continued to be all listed (at least according to my supervisors...). But the party continues. Scroll down for the next examples.

(Milinski et al. 2006)

(Milinski, Semmann, Krambeck & Marotzke 2006)

(Delbecq, Van den Ven & Gustafson 1975)

(Arentsen, Bressers & O'Toole 2000)


Arentsen, M.J. et al. 2000. Institutional and Policy Responses to Uncertainty in Environmental Policy: A Comparison of Dutch and U.S. Styles. Policy Studies Journal, 28(3), pp.p597, 15p.

Delbecq, A.L. et al. 1975. Group Techniques for Planning. A Guide to Nominal Group and Delphi Processes. Glenview, Illinois.: Scott, Foresman and Company.

Milinski, M. et al. 2006. Stabilizing the Earth's climate is not a losing game: Supporting evidence from public goods experiments. Proceedings of the National Academy of Sciences, 103(11), pp.3994-3998.

------------------------------------------------------------------------------

Example 3. Changed Zotero's Prefs to Harvard format 3. Bibliography puts the full authors' list back again, but the Milinski references keep having the dual personality:

(Milinski et al. 2006)

(Milinski, Semmann, Krambeck & Marotzke 2006)

(Delbecq, Van den Ven & Gustafson 1975)

(Arentsen, Bressers & O'Toole 2000)


Arentsen, MJ, Bressers, HTA & O'Toole, J 2000, “Institutional and Policy Responses to Uncertainty in Environmental Policy: A Comparison of Dutch and U.S. Styles..” Policy Studies Journal, vol. 28, no. 3, pp. p597, 15p.

Delbecq, AL, Van den Ven, AH & Gustafson, DH 1975, Group Techniques for Planning. A Guide to Nominal Group and Delphi Processes., Scott, Foresman and Company, Glenview, Illinois.

Milinski, M, Semmann, D, Krambeck, H & Marotzke, J 2006, “Stabilizing the Earth's climate is not a losing game: Supporting evidence from public goods experiments.” Proceedings of the National Academy of Sciences, vol. 103, no. 11, pp. 3994-3998.

----------------------------------------------------------------------------------

Example 4. Changed Zotero Prefs to Harvard format 7. First observation: all references went back to the "et al" format, but... behold: Second observation: the bibliography has now two articles with full authors' names and one with "et al". Why oh why...?


(Milinski et al., 2006)

(Milinski et al., 2006)

(Delbecq et al., 1975)

(Arentsen et al., 2000)


Arentsen, Maarten J; Bressers, Hans Th. A; O'Toole, Jr. (2000): „Institutional and Policy Responses to Uncertainty in Environmental Policy: A Comparison of Dutch and U.S. Styles.“. In: Policy Studies Journal. 28 (3), pp. p597, 15p.

Delbecq, Andre L; Van den Ven, Andrew H; Gustafson, David H (1975): Group Techniques for Planning. A Guide to Nominal Group and Delphi Processes. Glenview, Illinois.: Scott, Foresman and Company.

Milinski, Manfred et al. (2006): „Stabilizing the Earth's climate is not a losing game: Supporting evidence from public goods experiments“. In: Proceedings of the National Academy of Sciences. 103 (11), pp. 3994-3998.

-------------------------------------------------------------------------------

The whole thing just baffles me... It wonder if the bug is in individual references and not Zotero itself, eg, something might be wrong with Milinski, but I have looked at the three in Zotero and there is nothing that looks wrong with them. And they do all show up fine, only at different times. Now, if this is the case, and considering that most of my articles were automatically filled through Zotero's autocomplete, I have no idea how many more articles I might have that will cause situations like this (and I have hundreds of them).
The whole thing is just to coherent and unreliable, and sadly this means I have to go to some other software, since I'm the last stages of my PhD thesis writing and can not afford all this bugs right now...

Unless... someone could shed any light on this, see where the problem is, turns out to be something really simple, and it all ends up well.

Thanks for any help!
  • edited August 22, 2008
    I'm not affiliated with the Zotero development team, but I have recently been trying to write/alter some styles, and I hope I can give you some answers.

    I looked into some of your examples, and the 'bugs' you encounter all just seem to be things specified in the styles you were using. Perhaps a thing to stress is that the reason for the presence of multiple Harvard styles is that different universities seem to have adopted some differing implementations of the Harvard standard. Style 1 lists http://libweb.anglia.ac.uk/referencing/harvard.htm as the source of the reference documentation, style 2 http://www.leeds.ac.uk/library/training/referencing/harvard.htm , and style 3 http://library.curtin.edu.au/referencing/harvard.pdf . So I do not know which of these slightly differing styles suits your needs. Does your university provide specifications for the Harvard format? These specifications might be the same as those of one of the universities listed above. If not, a new style is required.

    Furthermore, as most of these styles don't seem to have been tested very thoroughly, there might also still be some errors in them. For instance, for style 1, the documentation specifies that "Where there are several authors (more than two), only the first author should be used, followed by ‘et al’ meaning ‘and others’: Green et al (1995)". However, style 1 is written such that only for papers with more than 3 authors 'et al' is used for the citation (the 'short reference group'), so only the Milinski paper qualifies for 'et al'. As style 1 is curated by the Zotero team, I can't fix this error myself, but one of the Zotero developers might be able to. It is rather simple:

    in http://www.zotero.org/styles/harvard1

    <citation>
    <option name="et-al-min" value="4"/>
    <option name="et-al-use-first" value="1"/>
    <option name="et-al-subsequent-min" value="4"/>
    <option name="et-al-subsequent-use-first" value="1"/>

    Here the 4's should be 3's. If you need this fix fast you could also do it yourself:
    http://forums.zotero.org/discussion/3138/?Focus=13497#Comment_13497

    I haven't checked the other styles yet for any errors, so I don't know if they show the expected behaviour.
  • As style 1 is curated by the Zotero team, I can't fix this error myself, but one of the Zotero developers might be able to.
    All the styles have open SVN commit access, including the built-in styles, so you're more than welcome to make the change yourself. Committed styles just automatically become dev styles until we mark them as public.

    It's also quite possible that Harvard 1 shouldn't be the default Harvard style. If any experts on Harvard style want to make a recommendation here, we'd be happy to hear it.
  • edited August 23, 2008
    I wasn't aware of that. I just committed the fix. Ocoiso, you can install the corrected Harvard 1 style by using this link:

    http://www.zotero.org/styles/harvard1/dev/install

    I'm not sure however what the style wants for journal articles in the bibliography itself. As far as I can see it only mentions books with multiple authors (here indeed "et al" should be used in case of more than 4 authors, and only the first author should be included).

    I just noticed that the repository gives an error for the updated style in the preview panel ("Error generating preview"). Am I doing something wrong, or does it just take some time to generate the preview?
  • I just noticed that the repository gives an error for the updated style in the preview panel ("Error generating preview"). Am I doing something wrong, or does it just take some time to generate the preview?
    The server we use to generate previews is currently unavailable. (I'll respond to your other post about that once it's back up.)
  • Rintze, thank you so much for your help and for the work you had in changing the style.
    It is indeed extremely frustrating that every Uni adopts its own version on Harvard. I'm based at Leeds Metropolitan, which although is right next door from Leeds Uni seems to also have adopted a different format for Harvard. I'm attaching the link below to our Harvard regulations: http://skillsforlearning.leedsmet.ac.uk/documents/harvard_2004.pdf

    One of the first and more obvious ones is once again the minimum number of authors before you put "et al." Leeds Met has decided it should be three. That has in fact surprised me because I was convinced it should be two and I don't remember seeing many examples with 3.

    Another example is the presentation of the year in the bibliography, that Leeds Met considers it should be in brackets and Leeds Uni doesn't. From the examples of my first post, only the German Harvard (format 7) is doing this (but then has those annoying quotation marks in the article/book title).

    I think the work you did with Harvard 1 was right on target, because that is in fact the Harvard presentation I am more used to see (and the most intuitive one). Unfortunately, the format I need is a mixture of all the available 4 and none of those in particular.

    One thing I could not understand from none of different the Harvard documentations is what do you do in the bibliography when you have 4 or more authors. All of them skip that subject. If I understand correctly from what you wrote, are we supposed to just write "et al" and include all names when we have 3 or less authors? I would consider that most unusual, and no quick examples are coming to my mind of finding that in published materials (though there might be). And it seems my supervisors also insist on not using "et al" at all in the bibliography.

    Finally, the one strange thing that keeps happening with the examples I provided, is the "dual personality" of Milinski et al, that both in Harvard 2 and Harvard 3 appears once with "et al." and once without. Zotero did that automatically, so is that a bug or am I to understand that there are citation formats where, for more than three authors, the first time you present a reference is with "et al" and after that you put all names? That would be really strange.

    Is it anywhere on the roadmap of zotero the plan to build a simple graphical way for the end-user to create customized style, like Endnote allows? That would sort a lot of style issues and removed some weight of the endnote team and valuable collaborators and volunteers.

    Once again, thanks for everything.
  • One of the first and more obvious ones is once again the minimum number of authors before you put "et al." Leeds Met has decided it should be three. That has in fact surprised me because I was convinced it should be two and I don't remember seeing many examples with 3.
    So in this regard the fixed Harvard 1 style is correct, right?
    Another example is the presentation of the year in the bibliography, that Leeds Met considers it should be in brackets and Leeds Uni doesn't. From the examples of my first post, only the German Harvard (format 7) is doing this (but then has those annoying quotation marks in the article/book title).
    This indeed seems to be a difference between Leeds Met and Uni. I guess this warrants a new style. I just uploaded one, now available in the style repository (http://www.zotero.org/styles, look for "Harvard - adapted for Leeds Met"). I also changed some things in the style I spotted in your guide.
    One thing I could not understand from none of different the Harvard documentations is what do you do in the bibliography when you have 4 or more authors. All of them skip that subject. If I understand correctly from what you wrote, are we supposed to just write "et al" and include all names when we have 3 or less authors? I would consider that most unusual, and no quick examples are coming to my mind of finding that in published materials (though there might be). And it seems my supervisors also insist on not using "et al" at all in the bibliography.
    Your guide does actually discuss this topic (see page 46): "Et al. (Abbreviation of the Latin term et alii meaning ‘and others’). This is commonly used as an abbreviation for ‘and others’ when there are more than four authors." The way I read this is that it is allowed to use et al. in the bibliography, but it's not a requirement. So you could also choose to include all authors, which would of course require a style change. As you indicated you would like to include all authors, I specified this in your new style.
    Finally, the one strange thing that keeps happening with the examples I provided, is the "dual personality" of Milinski et al, that both in Harvard 2 and Harvard 3 appears once with "et al." and once without. Zotero did that automatically, so is that a bug or am I to understand that there are citation formats where, for more than three authors, the first time you present a reference is with "et al" and after that you put all names? That would be really strange.
    This behaviour is specified in the styles. For Harvard 2 for example, the citation code includes the text:
    <option name="et-al-min" value="4"/>
    <option name="et-al-use-first" value="1"/>
    <option name="et-al-subsequent-min" value="6"/>
    <option name="et-al-subsequent-use-first" value="1"/>"
    So the first time a source is cited, et al. is used in case of 4 or more authors, while for any subsequent citation the bar is raised to 6 authors. I didn't check the style guide whether this is the required behaviour for Harvard 2, but the 'bug' you describe is exactly what you would expect from the style.
  • Rintze... you're magical! I cannot tell you how much I appreciate what you just did.
    Although I have been using Zotero to store and manage my references for more than a year now, this is the first time I have used the forum and I'm blown by the support and your swift and helpful contributions.
    I've just told my office mate about it and about the new "Harvard - adapted for Leeds Met" and he has decided to change to Zotero too! None of us likes very much working with Endnote (and we both have the latest version...).

    Thank you also for the clarifications about the multiple authors (that was indeed a surprise, i.e. that the number of referenced authors can change).
    I was already a Zotero "evangelist" for the students and colleagues here, but now that we have our own Leeds Met style I will be spreading the word even more!

    Cheers and all the best!
  • edited August 28, 2008
    You're welcome. Just let me know when you find any more differences between the style and the guide. It will probably still be a while before a graphical editor becomes available, and I don't mind digging into the Citation Style Language a bit more.
  • Hi Rintze,
    Many thanks for starting the ‚Harvard adapted for Leeds Met style’. I will have to use it for my work and take the freedom to point you towards some differences between the style guide from Leeds Met University’s website (SG)
    http://skillsforlearning.leedsmet.ac.uk/documents/harvard_2004.pdf
    and the style you created as currently in Zotero (SZ). I will refer to the Leeds Met style guide (SG) by pages:

    1) p5 'citing within your text'
    SG wants (author, year), currently not so in SZ (author year)
    for p.6 'direct quotations’: (author, year, p.), currently in SZ its (author year, p.)

    2)p12 'book in bibliography'
    SG wants the title (3) followed by a full stop; currently in SZ: title followed by comma,
    SG wants the place of publisher (6) followed by comma; currently in SZ ‚:’

    3) p13 ‚in:' reference
    SG wants ‚In:’ (4), currently in SZ ‚In’

    4) p15 ‚translations’
    SG wants 'Orgininal Language' and 'Translator' mentioned after title (4) – (7), currently in SZ the translator is not mentioned in bibliography. (Original language might be tricky as it is not usually a field in the zotero database, might have to do that manually?)

    5) p16 theses and dissertations
    SG wants thesis type (phd, mphil, etc) and university mentioned after title. SZ does not produce this information for theses entries in the bibliography.

    6) p38ff electronic sources
    SG wants the word [Internet] behind the title of web sites, web pages, e journal articles, e books; SZ currently doesn’t do that.
    For email discussion lists (p43) SG wants [Internet Discussion List], but again, this might be easier to adapt manually.

    Generally SG wants ‘Available from:’, currently SZ: ‘Available at’.

    Also SG wants [accessed day month year], currently SZ does [accessed month day, year]

    That’s what I found so far. Thanks very much for the support.
  • edited February 6, 2009
    I updated the style. I didn't touch point 4, as it is not possible to call Zotero's language field from a CSL style. For 5, enter the type of thesis (PhD, etc) in the Type field. For 6, I'm not sure how you make clear in Zotero that a certain source like a book is an "e-source", so (for now) the addition of the [Internet] label is limited to webpages. Also, Zotero doesn't really have a dedicated email discussion list-item, so that was a no-go as well. The other errors should be fixed.
  • Thanks for that!

    Is the updated version on the zotero style repository?

    I could only see the old version.
    http://www.zotero.org/styles/harvard-leeds-met/dev
  • edited February 8, 2009
    It should be, but it isn't. I guess the Style Repository hasn't (automatically) updated. For now you can install the updated style via:

    https://www.zotero.org/svn/csl/harvard-leeds-met.csl

    (download and drag the csl file into an open Firefox window)
  • Great, now it works!
    Thanks.
  • Hello again,
    I noticed one more problem with the zotero style 'harvard-leeds-met'.
    In the case of 'more than one reference at the same point in the text' (p.8 of the Leeds Met Style guide) the comma between author and year has to go:
    Example: (author year; author year) rather than (author,year; author, year).

    Many thanks for fixing that.
  • I don't think that is possible (yet) in CSL. I think we need the Bat-signal to get Bruce's attention.
  • This is one of those "stupid rules that's too complicated to implement." I see no reasonable way to do it, and I see no logical basis on which that style guide includes that example.

    Grumble, grumble ...

    On the practical level, I'd suggest a) contacting the style authors and tell them to fix this, and b) don't worry about it. If someone complains, you can fix it at the end.
  • Ah, I see. Will forward this to the style authors.
    With more and more people using Zotero here at Leeds Met, there is surely a case for fixing the style guide.
  • Hi there
    I have two more questions concerning the in-text reference with the word plug-in:
    1) In-text references are currently displayed in the font Times New Roman, size 12 in the word-document. Is there a way to change the default to Arial, size 11, the format I need to put the thesis in?
    2) if I add suffixes and prefixes to the in-text reference, the word-plug-in renders them to CAPITAL letters in the word document (this might be not style related, but rather word-plugin related, I suppose)

    Can someone help?

    Thanks
  • I found a solution to the first issue, thanks to ocoiso. In case someone comes across the same problem:

    The zotero word plug-in will insert using the default font set in word itself. Therefore change the default font in word itself (in the format menu go to font and default) and then zotero follows suit.
Sign In or Register to comment.