[Refdb-users] AUTHORLIST's ALTERNATETEXT and ALTERNATESTYLE attributes not producing correct output

[Bruce_D_Arcus]

Damn, another email I just lost. Anyway …

David wrote:

Some time ago support for alternative author text was added. Many
bibliographic styles use the pseudo-author ‘Anonymous’ when there is
no known author for a work. Other styles, however, use the title in
place of author when the author is missing. The ALTERNATETEXT and
ALTERNATESTYLE attributes for AUTHORLIST enable the user to specifiy
an alternate field to use for author when the author is missing.
ALTERNATESTYLE provides for the alternate text to be rendered in a
style different to that used for ‘regular’ authors – most such styles
would render title-as-author in italic.

I’m curious about this question as I’m tackling it for my CSL schema.
I actually think it’s rather tricky.

Observations:

1. What here is called “author” is in fact two different things.

   a. a creator (broader than author) name
   b. a sort key

2. While for almost all reference types the creator = the sort key,
   there are exceptions. Consider, for example, a legal case:

   Doe v. Jones (2002) …

“Doe v. Jones” is in fact a title, so that becomes what I call in my
system a “nonname-substitute.”

Here’s the trick part though:

Structurally-speaking, legal cases share the same structure as a
periodical article. However, if a name is missing on a newspaper
article, then the periodical title usually substitutes for it.

OK, so I was then thinking maybe the specification of this
redirect/substitution should not happen in the author/name list
configuration, but rather in the main layout.

So I have stuff like this currently:

   <reftype name="book">
     <creator>
       <names noname-substitute="Anonymous"/>
       <role>
         <prefix> </prefix>
       </role>
     </creator>
     <date>

… and:

   <reftype name="article">
     <creator>
       <names  noname-substitute="periodical-title"/>
     </creator>
     <date>

But this is somewhat awkward because the same logic ought to apply to
the citation mark too! Nevermind mind even that using attributes to
indicate both CDATA and pointers to nodes seems like a hack.

One possibility is a new top-level element called “sort-key”.

In any case, my goal all along has been a schema that is easy to
represent in a GUI, and is as simple as possible to get the necessary
functionality. Here’s an example (without the added feature). I’m a
bit stuck, and would appreciate suggestions.

<?xml version="1.0" encoding="utf-8"?>

author-year

Bruce D’Arcus

2004-07-30
2005-02-11

Chicago
15

geography
sample style file for author-year class





Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec





p.
pp.




¶
¶¶




§
§§







Ed.
Eds.




tran.
trans.





PhD Dissertation


letter




CD-ROM




(



,



:


)











(
)



,




:



,


(
)



,


,













(
)




, In





,




,
,



(


:

)







,


,










(
)



,



,










,



(
)





,


(
),


available from:

,


,









(
)




,












,


(
)





,


(
),


available from:

,


,









(
)



,



,



,

Bruce

[Bruce_D_Arcus]

David – the ticket I added is referring to this issue; I forwarded the
note to the list last week. The basic problem is that I can’t get
legal cases the correctly render now. Addressing this correctly has
the potential to make CSL significantly more complex, which I’d like to
avoid if possible.

Observations:

1. What here is called “author” is in fact two different things.

   a. a creator (broader than author) name
   b. a sort key

2. While for almost all reference types the creator = the sort key,
   there are exceptions. Consider, for example, a legal case:

   Doe v. Jones (2002) …

“Doe v. Jones” is in fact a title, so that becomes what I call in my
system a “nonname-substitute.”

Here’s the trick part though:

Structurally-speaking, legal cases share the same structure as a
periodical article. However, if a name is missing on a newspaper
article, then the periodical title usually substitutes for it.

OK, so I was then thinking maybe the specification of this
redirect/substitution should not happen in the author/name list
configuration, but rather in the main layout.

So I have stuff like this currently:

   <reftype name="book">
     <creator>
       <names noname-substitute="Anonymous"/>
       <role>
         <prefix> </prefix>
       </role>
     </creator>
     <date>

… and:

   <reftype name="article">
     <creator>
       <names  noname-substitute="periodical-title"/>
     </creator>
     <date>

But this is somewhat awkward because the same logic ought to apply to
the citation mark too! Nevermind mind even that using attributes to
indicate both CDATA and pointers to nodes seems like a hack.

One possibility is a new top-level element called “sort-key”.

In any case, my goal all along has been a schema that is easy to
represent in a GUI, and is as simple as possible to get the necessary
functionality. Here’s an example (without the added feature). I’m a
bit stuck, and would appreciate suggestions.

<?xml version="1.0" encoding="utf-8"?>

author-year

Bruce D’Arcus

2004-07-30
2005-02-11

Chicago
15

geography
sample style file for author-year class





Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec





p.
pp.




¶
¶¶




§
§§







Ed.
Eds.




tran.
trans.





PhD Dissertation


letter




CD-ROM




(



,



:


)











(
)



,




:



,


(
)



,


,













(
)




, In





,




,
,



(


:

)







,


,










(
)



,



,










,



(
)





,


(
),


available from:

,


,









(
)




,












,


(
)





,


(
),


available from:

,


,









(
)



,



,



,

Bruce

[Bruce_D_Arcus]

Here’s an idea

1. Define the sort-key explicitly, in the bibliography layout:

   
   
   
   
   
   
   
   
   
   
   
   …

2. Then use that to figure out how to format the citation (they
   wouldn’t ever be different; right?):

   ( , : )

It’s simple. Will it work? A GUI might be a little tricky.

Also, it’s a little awkward to include the role in the key, because
it’s really a separate issue. Yet separating it out is also a bit
awkward; would result in:

     <sort-key>
       <creator>
         <names/>
       </creator>
       <alternate use="container-title"/> <!-- content of element 

could be text -->

Blah.

Bruce

[M_David_Peterson1]

Hey Bruce,

When I get back later this evening I will run through this and see
what can be done. I’m rushing to get ready for my presentation at
noon at the moment but will have a good chunk of time this evening to
run through it.

Cheers

<M:D/

[Bruce_D_Arcus]

Just to flag my last post on this before tackling it David. The
previous one in this thread is more explanation of the issue.

Bruce