It's up - I did some cleaning up in the style but only two actual changes: I fixed the access macro (currently for anything with an accessed date but no URL you'd get the whole macro but with empty <>) and I added a collapse option to the citation, so that two works by the same author and year would display correctly as (Smith 1776a, b) Thanks for contributing
I don't understand the preview errors myself - have to ask, but no, it's not a problem, the style works.
If you want to update it, just post here with a link to the updated style on github. Remember to work on the version of the style that's in the repository, not the one you submitted.
on the previews see here
http://forums.zotero.org/discussion/18544/error-generating-preview/#Item_1
should work now and was never a problem with the style.
I am not sure if this is the correct place to post this question - if not I can move it to a new place.
The Harvard Anglia Ruskin University 2010 citation style does not appear to sort bibliographic entries according to publication date. It is required by the Anglia Ruskin style - according to the web site "where there are several works from one author or source they should by listed together but in date order, with the earliest work listed first."
Is this something that can be included in the citation style or is it something that has to be done manually?
I have had a brief look at the citation style and on line 134 in the <sort> section it looks as though this style is set to sort on book title: <key variable="title"/>.
When I changed this to the sort order found in the Chicago author-date style <key variable="issued"/>, the bibliography was sorted correctly in date order. However, this solution only worked when the Author name field(s) in the Zotero database are exactly the same for each work of a given author - and it seems to be important that they are exactly the same. So if I have 5 works by an author, then each record must have the same layout for author name - for example Good, M.P. (Good, M. P. - with space between M. and P. appears to be recognised as a different author).
There may be ways round this, but at least this works for now.
I would be grateful if my "reasoning" could be checked - and to make sure that other issues are not created by this change.
For info: in my new trial citation style the <sort> section, lines 132-135, reads: <sort> <key macro="author"/> <key variable="issued"/> </sort>
you're right, that's the way to go.
We usually don't sort by the date variable, but by the year-date macro (or whatever it's called in the style). The reason is that when author and year are the same, you want them sorted alphabetical by title, not by the (not displayed) month/day in which they were published. But that's obviously a rare issue.
You're also right on authors needing to be exactly the same for sort to work correctly.
When I changed <key variable="issued"/> to <key variable="year-date"/>, the sort order was not in the required date order. Perhaps this is not the way to write the code for this?
It looks to me as though this citation style already has code to automatically mark works published in the same year, allocating them yeara and yearb as dates (for example 2009a and 2009b); the <key variable="issued"/> then seems to sort on these altered dates. Is it sufficient to rely on this or is there a better way of sorting?
I do not know enough about XML coding to know what might produce the cleanest code. What would you advise?
you need key macro="year-date"
not key variable - there is no such variable.
Which item gets marked as 2009a and 2009b depends on their order in the bibliography.
As I say above, variable="issued" sorts according to the full date, including month and day.
Thanks for this - I understand now (I am "feeling my way" with the XML coding). I will update the citation style with your suggestion - and there are few other minor stylistic that I have noted. I will update those, test the new style - and then post a link to a new public gist here. It will be Anglia Ruskin 2011 version.
I presume you will be able to upload the style - and either update the 2010 version or delete it completely (whatever is the usual procedure). I will post again if I encounter any difficulties.
right - just post the style to github. Make sure that it validates and fulfills there requirements:
https://github.com/citation-style-language/styles/wiki/Style-Requirements
I have changed the <sort> section as suggested and added <key macro="title"/> at line 135 to ensure that works of the same date are sorted according to title (that was not the case until this line was added). I presume that is OK? If there is a better way, please advise.
The only other change is "Available from" to "Available at" in the web reference section (line 56)
The style validates OK and as far as I can see fulfils the requirements.
Thanks for your help on this - let me know if I need to make any further changes.
Just wondering if you managed to get a look at the edited code for the Harvard Anglia Ruskin citation style. You did say to remind you at the start of this renewed interaction...
Thanks for your help - it is great to get it sorted.
As before, there seems to be a problem with the preview function - I have noted the error with this style (and some others you may have been working on) on http://forums.zotero.org/discussion/18544/error-generating-preview/#Item_1
FYI - one of the things on my to do list is to rename the current Harvard Anglia Ruskin style Harvard 1 (replacing the current Harvard one - which is supposed to be based on Harvard Anglia Ruskin) and turn the A-R style into a dependent style.
Won't change anything from a user perspective, but don't be surprised if you find the style among the dependent styles.
The reason for getting involved in citations styles at all was because the Harvard 1 style was quite different from the Anglia Ruskin style. The current Anglia Ruskin citation style captures most of the current style requirements for its Harvard style documented on their web site, I think.
It is probably worth noting that I have not done extensive testing with the citation style, but with the most recent changes it produces the Anglia Ruskin format for the main type of bibliographical entries (books, journals, anthologies, web sites, web articles.
I am not entirely sure what a "dependent style" means, so perhaps the following question is not necessary... but let me ask it anyway.
Would there be any merit in replacing the Anglia Ruskin style completely and simply using Harvard 1? If so that would not be a problem as far as I was concerned provided we can keep Harvard 1 up to date with Anglia Ruskin formatting standards.
A dependent style consist merely of a link to an independent style.
When you install a dependent style, Zotero actually installs the independent style it points to, (but refers to it by the name of the dependent style).
The advantage of this is that we can have 100 journals, all using the same style, on the repository and if we need to fix or update the style we just need to modify one - i.e. the independent style.
The advantage of keeping Anglia Ruskin as a dependent style is that someone looking for Anglia Ruskin would find it immediately and wouldn't have to know that it's equivalent to Zotero's arbitarily named "Harvard 1"
Once you make the change, do I work on Harvard 1 style for furture updates or the Anglia Ruskin style?
If I remember correctly Harvard 1 is installed by default with Zotero - when you make the changes will that mean the new Harvard 1 (= Anglia Ruskin) will be installed by default?
I have been working with the updated Anglia Ruskin 2011 citation style and have discovered some issues with the style, particularly with a range of lesser used type of reference (magazine articles and newspaper articles have an "[online]" marker, interviews do not generate the correct bibliographical reference). I suspect there may be other issues as well.
In the current version of the style magazine/newspaper articles look as though they are coded for online articles by default, which means that there is no way to refer to hard copy references. Is there a better way to handle online references? The style for a website does not produce the required result - the current style for a website does not match Anglia Ruskin's style for online newspapers/magazines.
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?
I will test the citation style more fully this week to try to identify as many issues as I can. I will post my results here (unless I hear otherwise).
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? Perhaps when we know exactly what needs to be done the answer to that will be clearer...
Upgrade Storage
Filename: harvard-aru2010
If the citation style checks out what is the process for submitting it to the Zotero web site (assuming that is worthwhile and possible)?
I fixed the access macro (currently for anything with an accessed date but no URL you'd get the whole macro but with empty <>) and I added a collapse option to the citation, so that two works by the same author and year would display correctly as
(Smith 1776a, b)
Thanks for contributing
However I wasn't able to locate the style in the repository. Maybe I am missing something, so perhaps you could point me in the right direction...
And one other quick question... If I discover the style needs to be updated what needs to be done?
If you want to update it, just post here with a link to the updated style on github. Remember to work on the version of the style that's in the repository, not the one you submitted.
http://forums.zotero.org/discussion/18544/error-generating-preview/#Item_1
should work now and was never a problem with the style.
That will be very helpful
The Harvard Anglia Ruskin University 2010 citation style does not appear to sort bibliographic entries according to publication date. It is required by the Anglia Ruskin style - according to the web site "where there are several works from one author or source they should by listed together but in date order, with the earliest work listed first."
Is this something that can be included in the citation style or is it something that has to be done manually?
If anyone can advise on this I would be grateful
When I changed this to the sort order found in the Chicago author-date style <key variable="issued"/>, the bibliography was sorted correctly in date order. However, this solution only worked when the Author name field(s) in the Zotero database are exactly the same for each work of a given author - and it seems to be important that they are exactly the same. So if I have 5 works by an author, then each record must have the same layout for author name - for example Good, M.P. (Good, M. P. - with space between M. and P. appears to be recognised as a different author).
There may be ways round this, but at least this works for now.
I would be grateful if my "reasoning" could be checked - and to make sure that other issues are not created by this change.
For info: in my new trial citation style the <sort> section, lines 132-135, reads:
<sort>
<key macro="author"/>
<key variable="issued"/>
</sort>
We usually don't sort by the date variable, but by the year-date macro (or whatever it's called in the style). The reason is that when author and year are the same, you want them sorted alphabetical by title, not by the (not displayed) month/day in which they were published. But that's obviously a rare issue.
You're also right on authors needing to be exactly the same for sort to work correctly.
It looks to me as though this citation style already has code to automatically mark works published in the same year, allocating them yeara and yearb as dates (for example 2009a and 2009b); the <key variable="issued"/> then seems to sort on these altered dates. Is it sufficient to rely on this or is there a better way of sorting?
I do not know enough about XML coding to know what might produce the cleanest code. What would you advise?
not key variable - there is no such variable.
Which item gets marked as 2009a and 2009b depends on their order in the bibliography.
As I say above, variable="issued" sorts according to the full date, including month and day.
I presume you will be able to upload the style - and either update the 2010 version or delete it completely (whatever is the usual procedure). I will post again if I encounter any difficulties.
https://github.com/citation-style-language/styles/wiki/Style-Requirements
I have updated the title to Harvard - Anglia Ruskin University 2011 (this reflects the latest edition of the published style - http://libweb.anglia.ac.uk/referencing/harvard.htm ).
I have changed the <sort> section as suggested and added <key macro="title"/> at line 135 to ensure that works of the same date are sorted according to title (that was not the case until this line was added). I presume that is OK? If there is a better way, please advise.
The only other change is "Available from" to "Available at" in the web reference section (line 56)
The style validates OK and as far as I can see fulfils the requirements.
Thanks for your help on this - let me know if I need to make any further changes.
Thanks
As before, there seems to be a problem with the preview function - I have noted the error with this style (and some others you may have been working on) on http://forums.zotero.org/discussion/18544/error-generating-preview/#Item_1
Won't change anything from a user perspective, but don't be surprised if you find the style among the dependent styles.
The reason for getting involved in citations styles at all was because the Harvard 1 style was quite different from the Anglia Ruskin style. The current Anglia Ruskin citation style captures most of the current style requirements for its Harvard style documented on their web site, I think.
It is probably worth noting that I have not done extensive testing with the citation style, but with the most recent changes it produces the Anglia Ruskin format for the main type of bibliographical entries (books, journals, anthologies, web sites, web articles.
I am not entirely sure what a "dependent style" means, so perhaps the following question is not necessary... but let me ask it anyway.
Would there be any merit in replacing the Anglia Ruskin style completely and simply using Harvard 1? If so that would not be a problem as far as I was concerned provided we can keep Harvard 1 up to date with Anglia Ruskin formatting standards.
When you install a dependent style, Zotero actually installs the independent style it points to, (but refers to it by the name of the dependent style).
The advantage of this is that we can have 100 journals, all using the same style, on the repository and if we need to fix or update the style we just need to modify one - i.e. the independent style.
The advantage of keeping Anglia Ruskin as a dependent style is that someone looking for Anglia Ruskin would find it immediately and wouldn't have to know that it's equivalent to Zotero's arbitarily named "Harvard 1"
Once you make the change, do I work on Harvard 1 style for furture updates or the Anglia Ruskin style?
If I remember correctly Harvard 1 is installed by default with Zotero - when you make the changes will that mean the new Harvard 1 (= Anglia Ruskin) will be installed by default?
In the current version of the style magazine/newspaper articles look as though they are coded for online articles by default, which means that there is no way to refer to hard copy references. Is there a better way to handle online references? The style for a website does not produce the required result - the current style for a website does not match Anglia Ruskin's style for online newspapers/magazines.
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?
I will test the citation style more fully this week to try to identify as many issues as I can. I will post my results here (unless I hear otherwise).
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? Perhaps when we know exactly what needs to be done the answer to that will be clearer...