Coordinating the work of Biblatex style authors

Question

I am not sure whether this is an admissible question. If it is not, please close it.

The problem

Biblatex opens up the realistic possibility of highly complex citation systems such as those used in my field, law. It is an area very poorly served by existing tools, and in theory an area where LaTeX could be a brilliant tool.

However, as any style author quickly finds out, the complexity of these systems requires innovation. We have to work with non-standard entrytypes. And we have to use existing fields unconventionally, and to make heavy use of custom fields. The same, I think, applies also in other fields of humanities. Quite a lot of work is being done by several people, independently.

It is obviously not a bug, but a feature, that the end result of these labours are styles that produce different output. Different legal traditions and humanities disciplines have their own preferences, and the whole purpose of is to abstract these preferences.

But it is or will be a bug if the use of different fields and entrytypes is so idiosyncratic that .bib files produced for one system are radically incompatible, even for sources, such as treaties, international cases, and EU cases, which are common to many different traditions. To take one tiny example, Tobias Schwan's juradiss package records, I think, the court that decided a case in the Author field, whereas my work in progress currently uses the Institution field. Either might be fine. But it will be a blot on the overall coherence of the system if some of us use one, and some the other, for it makes each bib file style-specific.

The question

Is there any existing mechanism by which this sort of decision could be discussed collectively, and existing solutions recorded, not in order to impose compulsory norms, but to help us avoid too Balkanised a system? If not, how could one be created, and what is the best way to have something lightweight and not-too-bureaucratic? As a lawyer, I'm depressingly unfamiliar with the tools for effective cooperation that others here have vast experience of.

Answer 1

Here's my attempt at bringing together the things that have been suggested so far. First of all, and to answer the question proper, we could try to reach people who are working on biblatex styles by:

1. Sending them emails directly and telling them that there is a project to harmonise things;
2. Making a public announcement about it on a blog (either the TeX SX blog or someone's, as offered by Joseph Wright) and possibly also on widely read mailing lists;
3. Perhaps also making a public announcement on a non-LaTeX blog or website, so as to reach people who work in the field but who are not active members of any TeX community (which is often the case with law and the humanities);
4. In all events, a wiki or a blog (with as many redactors as there are people interested) could be created to cover that precise topic, and would hopefully be ranked decently by search engines as there is little coverage of these topics. Perhaps such a website could be created for all biblatex styles and coordination projects, which would make it more visible, as it would be easier to make it known throughout the LaTeX world.

If the only aim is to offer guidelines for style authors, explaining which fields to use for which data, and I think the discussion can be handled by email; then we publish (wiki, blog, CTAN) a nice pdf document explaining the proposed nomenclature. We will probably want to get in touch with the biblatex maintainers and coordinate that with them.

For instance, in the legal field, I feel that we would imperatively need name fields for the parties (plaintiff and defendant) because we may need to differentiate them, and we need a configurable \bibstring{contra} (for instance) between them. Note that biblatex 2.x can deal with arbitrary field names using custom data models.

While this might be the ideal solution for the humanities, in which (as far as I know) there are mainly university styles, I think we might want to be more ambitious regarding law.

Legal citation is very much codified and there are not that many different styles, however we need to handle the fact that most lawyers need to cite articles, laws and cases from different countries – all of which require a different style. Which is why I think we could perhaps aim at having a single package that deals with various languages based on the hyphenation field. If there are several different styles for the same language, we can declare them all as options, and pick a widely used one as the default. Hence, the different styles would be implemented based on:

1. the country (hyphenation field, based on babel language names);
2. the specific style, when it departs from the chosen default (using an \ifoption-like interface).

Some styles (such as the bluebook) have internationalisation in mind, so:

1. they can't be default;

2. if chosen, they should overwrite the default styles for other languages, using something like:

   \ifbluebook%
     …%
   \else%
     \iffieldequalstr{hyphenation}{british}{…}{}%
     \iffieldequalstr{hyphenation}{french}{…}{}%
   \fi
   

3. A similar system (but ultimately based on the institution's name instead of an option) can be used for things that have the same formatting in all languages (as, I think, most of the European Union law).

Such a project (which I think is what is ultimately needed if we want a working legal citation style) could be handled collaboratively through a version control system such as github or bitbucket (which are easy to set up and maintain). If we write a file that clearly separates the different country/international styles, it should be easy enough for the authors to maintain the part they are looking for.

Answer 2

(This is not an answer, but it won't fit a comment.)

I am very interested in this question, not only because I'm a minor LaTeX nerd, but also because I now have several law-citations.sty files in various folders and have been thinking about tidying everything up into a more comprehensive and stable file. I can't speak about modern legal studies, but legal history has its own set of peculiarities; and, although there are standards when it comes to citation, they are not widely known or practised outside this (tiny) field (especially by people who 'use' the sources but aren't really interested in legal history).

It seems to me, however, that any such 'package' must be developed with an eye on what biblatex is up to, if only because they are all citations and, as such, need to end up in the bibliography/notes. As I've suggested elsewhere (and here) citing pre-modern works is not ideally done with biblatex because of the various ways one might need to cite the work. But, these citations invariably need to fit into the way modern studies are cited, partly because they often include the information modern studies do (e.g., date and place of publication, especially if reprinted in facsimile or digitized as a .pdf).

One of the problems of a non-biblatex approach, though, is indexing, which makes me think there are other problems a real package would have to address (that haven't occurred to me). It can get very complex, very quickly, and so I would be interested to hear what others think about this. Take for example

Dig. 1.1.5 % or cited as D. 1.1.5 by pure Romanists (i.e., non-medievalists uninterested in canon law)

That seems simple enough, but very often you want to cite the 'Ordinary Gloss'; then the citation is (say):

Gl. ord. ad Dig. 1.1.5, s.v. 'dominia distincta'

where the actual citation is to a gloss on those specific words of Dig. 1.1.5. Usually, in at least one footnote, you need to mention the edition you used (e.g., Venice 1472). Or it might be someone's commentary, where you might see

Bartolus ad Dig. 1.1.5 nn. 14-16 (Basil 1560).

Here, the edition is paramount since these 'numbers' were added by later editors and are not necessarily identical across editions. (Earlier incunabula might not even have them.)

So far I haven't touched on the problem of canon law, which is actually more complex, but my question remains, I think: is it worth the effort to try to create something that works as a de facto (if not de iure) standard for citations, and how fine-grained should it be? (A \dig command might be simple enough; perhaps a \gloss command could be provided, which builds on commands like \dig; but, it seems to me, it is impossible to cater to every commentator without getting monstrosities that involve too many optional arguments.) I do all three, but only the first two are standard across files; the third is only defined if I'm using Bartolus in a specific paper.