Journal citation styles

Among the minor frustrations of publishing scientific papers is formatting, in particular of reference lists. Every journal has its own format, with only sister journals published by the same society generally using the same. Here just a sample of the same paper as formatted for the reference lists of four journals I have published in:

Breitwieser I, Ford KA, Smissen RD (2010) A test of reproductive isolation among three sympatric putative species of Craspedia (Asteraceae: Gnaphalieae) at Mt Arthur in New Zealand. New Zealand Journal of Botany 48, 75–81.

Breitwieser, I., Ford, K.A., Smissen, R.D. 2010. A test of reproductive isolation among three sympatric putative species of Craspedia (Asteraceae: Gnaphalieae) at Mt Arthur in New Zealand. New Zealand Journal of Botany. 48, 75–81.

Breitwieser I., Ford K.A., & Smissen R.D. (2010) A test of reproductive isolation among three sympatric putative species of Craspedia (Asteraceae: Gnaphalieae) at Mt Arthur in New Zealand. New Zealand Journal of Botany, 48, 75–81. 

Breitwieser, I., Ford, K.A. & Smissen, R.D. (2010) A test of reproductive isolation among three sympatric putative species of Craspedia (Asteraceae: Gnaphalieae) at Mt Arthur in New Zealand. New Zealand Journal of Botany 48: 75–81.

The problem is, of course, that it would be extremely fiddly to reformat every journal name into italics or every volume number into bold every time that needs to be done. Wouldn't it be nice if publishers agreed on a standard, at least across the same field of science?

This is where reference managers come in. Perhaps the best known currently is the commercial software EndNote, and I could get it at work. But because I want something that I can also use on my own computer at home I prefer to use the open source software Zotero, which works through the browser. It has got very user friendly and flexible, and as far as I can tell there are only two downsides left.

The first is that when I need an in-line reference I only have the options of inserting "(Smith 2010)" or "(2010)", the latter by ticking the 'suppress author' option. It would be neat if it could also offer the option of "Smith (2010)", as EndNote apparently does.

The second is quite simply that the number of available citation styles - see above - is still a bit limited. For example, the Zotero Style Repository does not currently appear to have:

• Botanical Journal of the Linnean Society
• Systematic Botany
• Taxon

These are not exactly obscure under any measure, they are three of the most important plant systematics journals globally. Admittedly the Bot. J. has the same one as the Biological Journal published by the same society, but Taxon for example has a very unique style.

There appear to be two ways of building one's own style. As they have to be coded in the CSL format, one can learn that format and 'program' a style. Obviously that is a bit of a barrier; it amounts to learning what is pretty much a super-specialised, extremely cryptic scripting language that the average scientist will not need for anything else, ever. I at least do not really consider learning all the commands an efficient use of my time.

Luckily, a Visual CSL Editor has been made available (apparently by Mendeley and Columbia University?), and yesterday I experimented with it a bit. The idea is to start by searching the style database of the site for an existing style closely matching the desired one, changing it as desired, and exporting it to disk as a CSL file. It can then simply be dragged into an open browser window to install it in Zotero (if Zotero is what you are using).

This is much easier than scripting a style but also has its hurdles to overcome. The thing looks like this:

In the upper right field we have the way the example citations would be formatted under the style, so that is clear.

Below that is a field showing all the current settings for the part of a reference that we are currently working on. This is where it gets a bit more difficult. In the above screenshot, we can guess that delimiter is what separates author names, and that prefix and suffix are elements that would precede or follow any author name. In other cases it is harder to figure out what something means.

But the main problem is navigation. Looking at the field on the left we see a bewildering tree of fields that may or may not need to be edited. What is more, many of them have the same name, for example "names". Here we realise, for example, that our guess from the previous paragraph was wrong. Although we are in a field called "names" (directly under another field called "names", nested two levels under yet another field called "names"!), the delimiters, prefixes and suffixes are actually for the collective of all authors of a paper. Why do they have to have a delimiter between each other when they are actually delimited by year of publication, name of paper, journal name, volume number, page numbers, and line break? God and the people who came up with the CSL format will know.

In other words, there are so many more fields that I can ignore that I find it difficult to find the ones I do need, and they have the same names and are often directly next to each other. On the other hand, I was unable to figure out how to tell the style that it should use "& al." instead of "et al.", as required by the journal Taxon. Maybe that is impossible, or maybe I just can't find the option?

Of course, this is available for free and much, much better than scripting. I am extremely grateful to have this option. But still, it is clear why few people would find themselves able and the time to write their own styles...