One solution to the magazine/newspaper issue would be to try to differentiate between online and hard copies. Could that be done based on the presence or absence of a URL in the database record?
yes, that's exactly the way to go - note that even when the item has a URL, if it has a page range Zotero will by default return false for <if variable="URL> - that makes a lot of sense e.g. for print journal articles retrieved from an online database - they have a URL, but you wouldn't want them treated like online sources
I also have a question as to whether it might be better to write some new code rather than try to fix the existing code?
that depends on a lot of things - how much there is to do, how quickly you code etc. - but there is most certainly no rule against rewriting from the scratch.
I have been doing some work on a further revision of the Anglia Ruskin style. I would value your thoughts on whether it is worth trying to re-order the current style file and attempt to get more of the code that is currently in the bibliography section into macros?
The reason for asking is that the bibliography section of the Open University Harvard style is generated exclusively from macros and text variables. However, the current version of the Anglia Ruskin Harvard style (and the similar Staffordshire University Harvard style) uses a combination of macros and other code for the bibliography. The Harvard 1 style looks as though it is somewhere between the two.
I assume (but that may be a bad assumption!) it is possible to produce a much more compact and ordered citation style file that the current one. I just wonder if it is worth the effort? I suppose if people know where to go to make any changes that will be needed in the future it may not matter too much.
Does the complexity of the file make any difference to its functionality?
the complexity of the file doesn't affect functionality, no.
Generally files that consist of concise bibliography sections made up mainly of macros are both faster to adapt/fix with future changes and produce more reliable results (e.g. for item types that aren't explicitly coded or missing data), so if you're working on this anyway, it might be worth it.
When I work to re-write and expand a style thoroughly that's certainly what I do.
Generally, every style can be written just with macros in the bibliography, but the more internally consistent a citation style is, the easier it is to do that - (i.e. if delimiters between different parts of a citation follow a consistent logic).
Thanks, that is helpful. Yes, I know what you mean as I had to add code to various sections of the citation style file to get it to work as I wanted.
I will test the file I now have to ensure that the citations and bibliography are as required by the Anglia Ruskin layout. I think I will release that file in the interim (to fix some of the inconsistencies with the current style on the Zotero repository). After that I will see if I can work to produce a more streamlined code file.
I will post a link to the code here when it is ready.
I have posted an updated citation style file at https://gist.github.com/1503919
It uses the current version of the Anglia Ruskin citation style and combines it with elements of the Staffordshire Uni version. As a code file it is fairly messy, especially in the bibliography section. However, with the testing I have done it works for most of the main elements of Anglia Ruskin Harvard (and is closer to the required style than the current version).
I would be grateful if the citation style repository could be updated with this code.
I am also working on a new citation style file. If everything goes according to plan it will be based more fully on macros and the bibliography section should be much more streamlined. It might be ready in the New Year if testing goes OK...
I had some time to look over the code. First, it's not valid csl - you should run the validator over the code - I think all of the errors should explain themselves, but let us know if you have any questions.
Then, I don't think the "extra" field (i.e. variable="note") should be used for styles on the repository except as annotation (as in: Chicago Manual of Style - Annotated and the like). People have all types of things in the extra field, some translators put information there, we don't want this used in general citation styles, even if it works for how you use it.
Finally, you want to test the style some with missing fields - e.g. I noted that journal articles that don't have issue and volume informations don't look right.
I am sorry, I appear to have posted an earlier version of the citation style by mistake (I think I must not have transferred the final version from the Zotero trial pane). I did check that the file I was working with verified OK - and I was pretty sure had fixed the errors noted above. Apologies for that - I do not mean to give you any extra work - I am sure you are busy enough without that!
However, I have done some work on your other comments and have checked the style for works with missing fields. I have also removed the variable="note" references - they were to help with some formatting/labelling (I found another way to achieve the same result).
The code is posted at: https://gist.github.com/1530871 (and I have checked I have posted the correct file this time!). This file verifies OK.
I have also been working on an entirely new citation style for the Anglia Ruskin style (as noted above). I have done a fair bit of initial testing and it seems be be working fine so far. At what stage would it be worth releasing the file for further inspection and testing?
thanks, I'll take a look asap.
Put up the new version once you think it's overall performance is better than the old one. Since I want to turn this into Harvard 1 I'll do some testing on it myself.
I have just realised that the citation style file I posted needs two minor changes... The details are (the line numbers refer to the style as posted):
1. There is a line missing at line 353> The following needs to be added: <text macro="online"/>
This should give the following result: <else-if type="article-journal"> <text macro="title" prefix=" " suffix="."/> <text macro="editor" prefix=" "/> <text variable="container-title" prefix=" " font-style="italic" suffix=", "/> <text macro="online"/>
2. Lines 298 and 299 need to be swapped to give: <else-if type="webpage"> <choose> <if variable="genre"> <group prefix=" " suffix=" "> <text variable="title" suffix=". "/> <text variable="container-title" font-style="italic" suffix="."/> <text macro="online-medium" suffix=" "/> </group> </if>
With these changes the style I posted now works fine with all of the references I am using (and the majority of the Anglia Ruskin samples I have tested). However, I know there are some shortcomings that we might be able to address in the new style.
I should have the new style ready for posting next week. The end results at the minute are roughly the same, however I think that alterations and additions are much more straightforward with the new file.
I have a number of questions about some of the fields I have used in order to generate some of the more complex layouts required. When I post the new file and the questions should I continue in this part of the forum or start a new conversation, perhaps under Anglia Ruskin 2011? It makes no difference to me - others may find this forum too long already.
I have been testing a new citation style file and have encountered what seems to be a problem with Zotero/CSL metadata field mappings. According to http://gsl-nagoya-u.net/http/pub/csl-fields/index.html a Zotero dictionary entry should map to CSL type "chapter". However in my tests this mapping does not appear to work as expected. Book sections are displayed as expected, but a dictionary entry is not treated as CSL "chapter". The same situation seems to be the same for Encyclopaedia article. In fact I can only get these entries to work as expected when I change "if type="chapter"" to "if type="chapter" match="none"".
I checked the Chicago author-date citation style, which does work for the dictionary and encyclopaedia entries, but only because "chapter" appears not to be specified in the "if type=" statements of the title macro.
Am I missing something here - or have the field mappings changed in some way?
Thanks for that. I guessed it must have been the case as it was the only way I could get things to work.
The new style is very close to being ready (it look longer than I had planned as I ended up re-working the whole style file). I need to check through it for a final time and then I will post it for further testing/comment - that should be by the weekend.
For some of the less common layouts I have had to code some things in a somewhat non-intuitive fashion. I have a list of these that I could post here (if that is OK for a forum post) - you may decide that Harvard1 may need to be slightly simpler (or that the way I have done thing may have knock-on effects that I am not aware of). Anyway you can see what you think when I post the file - I presume you will be able to do some testing with your data (as far as I can see it works fine with my data set).
yes, please post the list here and yes, I'll want to make sure that Harvard 1 doesn't do unusual/unexpected things, so I might (!) edit those out, depending on what exactly they are.
Anglia Ruskin (AR) Harvard CSL file comments/issues/questions:
Layout details are here: http://libweb.anglia.ac.uk/referencing/harvard.htm
In my comments below I have noted the sections (in brackets) where the required layouts can be found (pasted references lost formatting).
In order to achieve the required layouts I have had to code some "workarounds". These may not be necessary for a more straightforward Harvard scheme, but they do produce the layouts that AR requires. (However, it would be great if Harvard1 was as close as possible to AR, especially if Harvrd 1 is included by default in Zotero download packages.)
I realise there may be better ways of coding for these items. I am also aware that some of the ways I have achieved these results are not always intuitive. I will simply highlight these issues and if we need to discuss/negotiate/modify we can. Some of them may not affect overall performance and can be left in place, but you will be aware where these things could have adverse knock-on effects.
Some of the macros in the file could be combined, but at this stage I thought I would leave them as they are. You may have an opinion on what should be done.
Journal/magazine articles - code lines 332-367
Journal articles - these will not record a date (apart from initial year date) if there is any volume information present; if none is present a date is inserted (if it is available).
Magazine articles - a date (where present) will always be inserted (and volume and issue where present, but this is not formatted in the same way as a journal). If people want one format as opposed to the other for a specific article they could change the record type.
It is not completely clear from the AR style guide whether month dates should be in short form or in long form. I have, therefore, decided to have all of the month dates appear in long form (but there may be a case for magazine and newspaper dates to be in short form). This matched the format for access dates. Any thoughts?
Journal from a database (Journal from a database) - code lines: 446-456
I have used the "series" field of the "Journal article" record for the database details. The CSL file contains the AR required "Available through", but if this is a problem I presume it could be entered in the actual data field.
The [e-journal] marker needs to be manually added to the "volume" field in the Zotero record.
Perhaps you can advise.
If the reference is to a journal abstract from a database the requirement is that "Abstract only" is inserted. I did not want to use the "Abstract" field as it may cause other problems. I presume it could be entered manually if it were needed (and anyway I do not know how often this sort of citation is needed).
Manuscript/PDF doc (Archive material)
I have used the Zotero "type" field as the way to produce the document genre "[manuscript]".
The "place" needs the actual location (London in the AR example) and the library; the "archive" field is the name of the collection.
Although "manuscript" could be used for a PDF document, most will probably use "Web Page". In that case the "Website Type" can be entered as PDF; this maps to a media descriptor to produce the required layout.
e-book (E-books and pdfs) - code lines: 259-261
To achieve the layout the [e-book] will be generated when the CSL "type" is a book and when there is a URL. This could in theory cause other conflicts, but it is unlikely for a book to have a URL unless it is an e-book.
Will this work?
Dictionary and encyclopedia articles - code lines
Clarified in post above.
I have presumed an "entry-encyclopedia" as well as "entry-dictionary"
Bill (Acts of parliament; Statutory instruments) - code lines: 161-175
For a bill I have used the "Code Volume" field to be used for place and publisher (there was no other fields for this). I have no experience of citing bills, so this may cause a problem. I think this needs clarified.
Interview (Interviews)
I have set up the "interview" type to allow it to be used for a radio interview. However, the interviewer details needs to be entered in the "archive" field and the radio programme details in the "abstract" field. Not a great solution, and may cause other problems.
This one may have to be left for manual input,
Real life interviews need noted in an Appendix, but this record type could be used if required.
The following items do not quite match the full Anglia Ruskin style, but they do come close.
Photo (Pictures, images and photographs) - code lines: 439-445
I have used the "abstract" field for the text entry required by AR style - probably OK in this instance.
I could not manage the layout for online images that do not have an author (Electronic images) - this is a strange layout. The nearest I could manage was if this sort of citation was entered as Item Type "web page" with the description of the image in the author field. It is not a good solution, but it works (and does not need any other coding than what exists for web page - unless you can do it!).
I can't imagine this being in high demand.
Emails (Email correspondence)
I have not included email addresses as per the AR style (are they really needed?); they could be added manually if required.
Legal cases (Law reports)
I have been able to code for UK/Ireland "neutral citations". Other citation styles had too many internal variables for me to understand, never mind code! But this gives at least a basic functionality. I see from the forums (http://forums.zotero.org/discussion/21467/style-request-legal-bibliography-entries-in-apa-6th/#Item_0) that I am not the only one!
Blog post (Blogs)
Because Zotero maps this to CSL "webpage" I was not able to generate the exact AR layout for a blog entry (I don't think it matters too much)
I presume that it is still possible for me to make changes if I discover any further issues. I will post them here anyhow.
I hope all of that makes sense - if not please let me know
I'll look through the issues in detail, but at this point my main concern is that a style with so many moving part is very likely to be fragile (n the sense of - produce wrong output when fields are missing) if you're relying on affixes - as opposed to groups and delimiters - as heavily as you do.
I am learning this as I go... so I was not aware of that issue. I have tried to test the file with a range of variable, including when fields are missing. However, if it will produce a more robust style then I would be happy to look at the issue of groups and delimiters and see what can be done.
See what you think overall and let me know if it would be worth more work on my part on this.
Hey Steve - just to say I'm sorry I haven't gotten to this - I looked at it, but it's tricky - the style guide is a total mess (especially the treatment of articles from databases - no one actually does what they ask for citing the specific database and article comes from) and I'm reluctant to put this up as a generic Harvard style - since my guess is that 95% of cited journal articles come from databases today, this is insanity:
"Boughton, J.M., 2002. The Bretton Woods proposal: an in depth look. Political Science Quarterly [e-journal] 42 (6) Available through: Blackwell Science Synergy database [Accessed 12 June 2005]."
It's also wrong. PSQ doesn't become an "e-journal" when accessed online. Is that really what they want?
I guess if we put this up as Harvard Anglia we should follow the guide, but then I'll base Harvard 1 on something more reasonable - Havard Limerick looks like it might be a good choice - apparently it's quite popular (doesn't exist yet - I'll code that soon).
The Bluebook(TM) requires similar treatment for law case items obtained from a commercial content provider, but only when no "print" version of the judgment is available. In the MLZ law styles, I've used abused the "archive" variable to hold the provider details. (Agree, though, that providing details in the cite of the channel through which the PDF of a published article happens to have been obtained is over the top.)
No problem about the delay, Sebastian. Yes I agree with you - I would be very unlikely to cite an article that way (or to ask any of our students to do it). We are following the guide (we do not have any links with Anglia Ruskin University), but I would not insist on that sort of detail for our students.
I am glad that it wasn't just me who thought that the official style guide was pretty complex!
The reason for including as many of the Anglia Ruskin options as I could was because I felt that if I was trying to produce a specific style it should reflect as much as possible of the official guidelines (which turned out to be more difficult than I had thought!). However, I would have no problem in providing a "simplified" Anglia Ruskin citation guide that simply dropped all of this more complex (and, perhaps, unnecessary) stuff. If an institution required the extra detail then I assume it could always be added manually to a Word document (and as I noted above some of the way in which it actually works is not what I would call intuitive).
I would be more than happy to work towards a "basic" Anglia Ruskin version (which may mean we could cut out most of the code that was for the more detailed AR citations - database, abstract from database, in fact most of the things that I have noted above).
And yes, I do have a vested interest! If a workable Anglia Ruskin style were available as part of the Zotero package that would make my job of helping students to get up and running with Zotero a little easier!
Let me know what you think - and if I need to do some more work on the citation code I am happy to do that
When you refer to Harvard Limerick do you mean the style guide published by the University of Limerick? If so they have a similar requirement for an article accessed via a database: http://www3.ul.ie/~library/referencing/harvard/artWebOnlyDb.html
Steven - yes - a "simpler" version would be nice - we could then just have one Harvard Anglia that's the "complete" version and Harvard 1 would be a simplified version based on it.
How about I give your style another look-over and then suggest what I would suggest taking out - I'd leave the final decision for you, but I figure it's be helpful for you to get a sense of what I think makes sense. E.g. I would like a lot of the hoops you jump through for legal citations in - there is no real downside to that.
Tthe Limerick styleguide only requires that type of citation for "Web Only" articles - which are currently still quite rare. If you look at the entry for journal articles they make clear that PDF versions of paper-journal articles are to be cited as regular journal articles - that seems reasonable and in line with common practice to me.
Sebastian, I am more than happy to work towards that. I had not anticipated some of the Anglia Ruskin complexities when I set out to prepare the new citation style (I have only ever had to use the more common types of citation). I think with hindsight I probably tried to be too comprehensive - but it was a good way to learn XML coding!
My guess is that we could easily live without quite a few of more complex aspects of the citation style (and some will probably never be used!) If you are happy to have a look and give me some pointers as to what you think needs to be done I am happy to do some more work. If we end up with a more usable and possibly more robust style then I think a simplified and less than comprehensive style for Anglia Ruskin will work fine. None of our students need anything very complex.
There are some things that I would like to keep in - like for instance the film and DVD citations (I have had a specific request for that). But see what you think - and let me know.
Yes, I see Limerick Harvard issue - though it is a little ambiguous. In the section on "Article - web only - accessed from a database" the guideline is: "If the article is retrieved from an aggregated database such as EBSCO, whether in PDF or HTML format, provide the name of the database and the date accessed." The "whether in PDF" is the reason for the ambiguity as in the main entry on journal articles it is clear that a PDF version is the same a paper version. But you could ignore the ambiguity if you decide to go with Limerick Harvard. Maybe we should ignore it for Anglia Ruskin too - and maybe even ask them to re-visit their style guide!
Here is another Irish university that has removed any ambiguity: http://www.library.dcu.ie/LibraryGuides/DCU%20Library%20Guide%20to%20Harvard%20Style%20of%20Citing%20&%20Referencing/player.html
Steven - OK, I had a quick look. I would suggest you take out the odd e-journal prescriptions. Also, since people use abstracts for descriptions of items, I suggest you remove everything that uses the abstract field. The only other messy thing are the law citations and I don't see any harm in leaving those in.
When I looked over the style, I noticed that you do a lot of testing for different author types (as in "if there's no author, print the editor" etc.). In most cases that's not necessary - use "substitute" instead - look at any number of styles for an example - that will make things much easier.
Thanks for that - I will make the changes you have suggested and I will see if there are other more obscure items that could be dropped. I don't know much about law citations (everything I know is in the citation style!) so I am happy to leave those alone!
And just to give you an idea of time-frame I should be able to have something ready for early next week.
Upgrade Storage
<if variable="URL>- that makes a lot of sense e.g. for print journal articles retrieved from an online database - they have a URL, but you wouldn't want them treated like online sourcesThe reason for asking is that the bibliography section of the Open University Harvard style is generated exclusively from macros and text variables. However, the current version of the Anglia Ruskin Harvard style (and the similar Staffordshire University Harvard style) uses a combination of macros and other code for the bibliography. The Harvard 1 style looks as though it is somewhere between the two.
I assume (but that may be a bad assumption!) it is possible to produce a much more compact and ordered citation style file that the current one. I just wonder if it is worth the effort? I suppose if people know where to go to make any changes that will be needed in the future it may not matter too much.
Does the complexity of the file make any difference to its functionality?
Generally files that consist of concise bibliography sections made up mainly of macros are both faster to adapt/fix with future changes and produce more reliable results (e.g. for item types that aren't explicitly coded or missing data), so if you're working on this anyway, it might be worth it.
When I work to re-write and expand a style thoroughly that's certainly what I do.
Generally, every style can be written just with macros in the bibliography, but the more internally consistent a citation style is, the easier it is to do that - (i.e. if delimiters between different parts of a citation follow a consistent logic).
I will test the file I now have to ensure that the citations and bibliography are as required by the Anglia Ruskin layout. I think I will release that file in the interim (to fix some of the inconsistencies with the current style on the Zotero repository). After that I will see if I can work to produce a more streamlined code file.
I will post a link to the code here when it is ready.
It uses the current version of the Anglia Ruskin citation style and combines it with elements of the Staffordshire Uni version. As a code file it is fairly messy, especially in the bibliography section. However, with the testing I have done it works for most of the main elements of Anglia Ruskin Harvard (and is closer to the required style than the current version).
I would be grateful if the citation style repository could be updated with this code.
I am also working on a new citation style file. If everything goes according to plan it will be based more fully on macros and the bibliography section should be much more streamlined. It might be ready in the New Year if testing goes OK...
First, it's not valid csl - you should run the validator over the code - I think all of the errors should explain themselves, but let us know if you have any questions.
Then, I don't think the "extra" field (i.e. variable="note") should be used for styles on the repository except as annotation (as in: Chicago Manual of Style - Annotated and the like). People have all types of things in the extra field, some translators put information there, we don't want this used in general citation styles, even if it works for how you use it.
Finally, you want to test the style some with missing fields - e.g. I noted that journal articles that don't have issue and volume informations don't look right.
However, I have done some work on your other comments and have checked the style for works with missing fields. I have also removed the variable="note" references - they were to help with some formatting/labelling (I found another way to achieve the same result).
The code is posted at: https://gist.github.com/1530871 (and I have checked I have posted the correct file this time!). This file verifies OK.
I have also been working on an entirely new citation style for the Anglia Ruskin style (as noted above). I have done a fair bit of initial testing and it seems be be working fine so far. At what stage would it be worth releasing the file for further inspection and testing?
Thanks again for your help.
Put up the new version once you think it's overall performance is better than the old one. Since I want to turn this into Harvard 1 I'll do some testing on it myself.
1. There is a line missing at line 353>
The following needs to be added: <text macro="online"/>
This should give the following result:
<else-if type="article-journal">
<text macro="title" prefix=" " suffix="."/>
<text macro="editor" prefix=" "/>
<text variable="container-title" prefix=" " font-style="italic" suffix=", "/>
<text macro="online"/>
2. Lines 298 and 299 need to be swapped to give:
<else-if type="webpage">
<choose>
<if variable="genre">
<group prefix=" " suffix=" ">
<text variable="title" suffix=". "/>
<text variable="container-title" font-style="italic" suffix="."/>
<text macro="online-medium" suffix=" "/>
</group>
</if>
With these changes the style I posted now works fine with all of the references I am using (and the majority of the Anglia Ruskin samples I have tested). However, I know there are some shortcomings that we might be able to address in the new style.
I should have the new style ready for posting next week. The end results at the minute are roughly the same, however I think that alterations and additions are much more straightforward with the new file.
I have a number of questions about some of the fields I have used in order to generate some of the more complex layouts required. When I post the new file and the questions should I continue in this part of the forum or start a new conversation, perhaps under Anglia Ruskin 2011? It makes no difference to me - others may find this forum too long already.
I checked the Chicago author-date citation style, which does work for the dictionary and encyclopaedia entries, but only because "chapter" appears not to be specified in the "if type=" statements of the title macro.
Am I missing something here - or have the field mappings changed in some way?
What's the status of the new style? I'd like to get it up asap.
The new style is very close to being ready (it look longer than I had planned as I ended up re-working the whole style file). I need to check through it for a final time and then I will post it for further testing/comment - that should be by the weekend.
For some of the less common layouts I have had to code some things in a somewhat non-intuitive fashion. I have a list of these that I could post here (if that is OK for a forum post) - you may decide that Harvard1 may need to be slightly simpler (or that the way I have done thing may have knock-on effects that I am not aware of). Anyway you can see what you think when I post the file - I presume you will be able to do some testing with your data (as far as I can see it works fine with my data set).
I will post the comments/issues/questions separately below.
I will be interested to see what you think.
Layout details are here: http://libweb.anglia.ac.uk/referencing/harvard.htm
In my comments below I have noted the sections (in brackets) where the required layouts can be found (pasted references lost formatting).
In order to achieve the required layouts I have had to code some "workarounds". These may not be necessary for a more straightforward Harvard scheme, but they do produce the layouts that AR requires. (However, it would be great if Harvard1 was as close as possible to AR, especially if Harvrd 1 is included by default in Zotero download packages.)
I realise there may be better ways of coding for these items. I am also aware that some of the ways I have achieved these results are not always intuitive. I will simply highlight these issues and if we need to discuss/negotiate/modify we can. Some of them may not affect overall performance and can be left in place, but you will be aware where these things could have adverse knock-on effects.
Some of the macros in the file could be combined, but at this stage I thought I would leave them as they are. You may have an opinion on what should be done.
Journal/magazine articles - code lines 332-367
Journal articles - these will not record a date (apart from initial year date) if there is any volume information present; if none is present a date is inserted (if it is available).
Magazine articles - a date (where present) will always be inserted (and volume and issue where present, but this is not formatted in the same way as a journal). If people want one format as opposed to the other for a specific article they could change the record type.
It is not completely clear from the AR style guide whether month dates should be in short form or in long form. I have, therefore, decided to have all of the month dates appear in long form (but there may be a case for magazine and newspaper dates to be in short form). This matched the format for access dates. Any thoughts?
Journal from a database (Journal from a database) - code lines: 446-456
I have used the "series" field of the "Journal article" record for the database details. The CSL file contains the AR required "Available through", but if this is a problem I presume it could be entered in the actual data field.
The [e-journal] marker needs to be manually added to the "volume" field in the Zotero record.
Perhaps you can advise.
If the reference is to a journal abstract from a database the requirement is that "Abstract only" is inserted. I did not want to use the "Abstract" field as it may cause other problems. I presume it could be entered manually if it were needed (and anyway I do not know how often this sort of citation is needed).
Manuscript/PDF doc (Archive material)
I have used the Zotero "type" field as the way to produce the document genre "[manuscript]".
The "place" needs the actual location (London in the AR example) and the library; the "archive" field is the name of the collection.
Although "manuscript" could be used for a PDF document, most will probably use "Web Page". In that case the "Website Type" can be entered as PDF; this maps to a media descriptor to produce the required layout.
e-book (E-books and pdfs) - code lines: 259-261
To achieve the layout the [e-book] will be generated when the CSL "type" is a book and when there is a URL. This could in theory cause other conflicts, but it is unlikely for a book to have a URL unless it is an e-book.
Will this work?
Dictionary and encyclopedia articles - code lines
Clarified in post above.
I have presumed an "entry-encyclopedia" as well as "entry-dictionary"
Bill (Acts of parliament; Statutory instruments) - code lines: 161-175
For a bill I have used the "Code Volume" field to be used for place and publisher (there was no other fields for this). I have no experience of citing bills, so this may cause a problem. I think this needs clarified.
Interview (Interviews)
I have set up the "interview" type to allow it to be used for a radio interview. However, the interviewer details needs to be entered in the "archive" field and the radio programme details in the "abstract" field. Not a great solution, and may cause other problems.
This one may have to be left for manual input,
Real life interviews need noted in an Appendix, but this record type could be used if required.
The following items do not quite match the full Anglia Ruskin style, but they do come close.
Photo (Pictures, images and photographs) - code lines: 439-445
I have used the "abstract" field for the text entry required by AR style - probably OK in this instance.
I could not manage the layout for online images that do not have an author (Electronic images) - this is a strange layout. The nearest I could manage was if this sort of citation was entered as Item Type "web page" with the description of the image in the author field. It is not a good solution, but it works (and does not need any other coding than what exists for web page - unless you can do it!).
I can't imagine this being in high demand.
Emails (Email correspondence)
I have not included email addresses as per the AR style (are they really needed?); they could be added manually if required.
Legal cases (Law reports)
I have been able to code for UK/Ireland "neutral citations". Other citation styles had too many internal variables for me to understand, never mind code! But this gives at least a basic functionality. I see from the forums (http://forums.zotero.org/discussion/21467/style-request-legal-bibliography-entries-in-apa-6th/#Item_0) that I am not the only one!
Blog post (Blogs)
Because Zotero maps this to CSL "webpage" I was not able to generate the exact AR layout for a blog entry (I don't think it matters too much)
I presume that it is still possible for me to make changes if I discover any further issues. I will post them here anyhow.
I hope all of that makes sense - if not please let me know
See what you think overall and let me know if it would be worth more work on my part on this.
"Boughton, J.M., 2002. The Bretton Woods proposal: an in depth look. Political Science Quarterly [e-journal] 42 (6) Available through: Blackwell Science Synergy database [Accessed 12 June 2005]."
It's also wrong. PSQ doesn't become an "e-journal" when accessed online. Is that really what they want?
I guess if we put this up as Harvard Anglia we should follow the guide, but then I'll base Harvard 1 on something more reasonable - Havard Limerick looks like it might be a good choice - apparently it's quite popular (doesn't exist yet - I'll code that soon).
I am glad that it wasn't just me who thought that the official style guide was pretty complex!
The reason for including as many of the Anglia Ruskin options as I could was because I felt that if I was trying to produce a specific style it should reflect as much as possible of the official guidelines (which turned out to be more difficult than I had thought!). However, I would have no problem in providing a "simplified" Anglia Ruskin citation guide that simply dropped all of this more complex (and, perhaps, unnecessary) stuff. If an institution required the extra detail then I assume it could always be added manually to a Word document (and as I noted above some of the way in which it actually works is not what I would call intuitive).
I would be more than happy to work towards a "basic" Anglia Ruskin version (which may mean we could cut out most of the code that was for the more detailed AR citations - database, abstract from database, in fact most of the things that I have noted above).
And yes, I do have a vested interest! If a workable Anglia Ruskin style were available as part of the Zotero package that would make my job of helping students to get up and running with Zotero a little easier!
Let me know what you think - and if I need to do some more work on the citation code I am happy to do that
When you refer to Harvard Limerick do you mean the style guide published by the University of Limerick? If so they have a similar requirement for an article accessed via a database: http://www3.ul.ie/~library/referencing/harvard/artWebOnlyDb.html
How about I give your style another look-over and then suggest what I would suggest taking out - I'd leave the final decision for you, but I figure it's be helpful for you to get a sense of what I think makes sense. E.g. I would like a lot of the hoops you jump through for legal citations in - there is no real downside to that.
Tthe Limerick styleguide only requires that type of citation for "Web Only" articles - which are currently still quite rare. If you look at the entry for journal articles they make clear that PDF versions of paper-journal articles are to be cited as regular journal articles - that seems reasonable and in line with common practice to me.
My guess is that we could easily live without quite a few of more complex aspects of the citation style (and some will probably never be used!) If you are happy to have a look and give me some pointers as to what you think needs to be done I am happy to do some more work. If we end up with a more usable and possibly more robust style then I think a simplified and less than comprehensive style for Anglia Ruskin will work fine. None of our students need anything very complex.
There are some things that I would like to keep in - like for instance the film and DVD citations (I have had a specific request for that). But see what you think - and let me know.
Yes, I see Limerick Harvard issue - though it is a little ambiguous. In the section on "Article - web only - accessed from a database" the guideline is: "If the article is retrieved from an aggregated database such as EBSCO, whether in PDF or HTML format, provide the name of the database and the date accessed." The "whether in PDF" is the reason for the ambiguity as in the main entry on journal articles it is clear that a PDF version is the same a paper version. But you could ignore the ambiguity if you decide to go with Limerick Harvard. Maybe we should ignore it for Anglia Ruskin too - and maybe even ask them to re-visit their style guide!
Here is another Irish university that has removed any ambiguity: http://www.library.dcu.ie/LibraryGuides/DCU%20Library%20Guide%20to%20Harvard%20Style%20of%20Citing%20&%20Referencing/player.html
When I looked over the style, I noticed that you do a lot of testing for different author types (as in "if there's no author, print the editor" etc.). In most cases that's not necessary - use "substitute" instead - look at any number of styles for an example - that will make things much easier.
And just to give you an idea of time-frame I should be able to have something ready for early next week.