How should I cite my old/new diversity preprint?
July 29, 2014
Recently, I published an old manuscript of mine as a PeerJ Preprint.
I wrote this paper in 2003-4, and it was rejected without review when I submitted it back then. (For, I think, specious reasons, but that’s a whole nother discussion. Forget I mentioned it.)
I haven’t touched the manuscript since then (except to single-space it for submission as a preprint). It’s ten years old. That’s a problem because it’s an analysis of a database of dinosaur diversity, and as everyone knows, the rate of recognising new dinosaurs has gone through the roof. That’s the reason I never made any attempt to update and resubmit it: dinosaur diversity is a fast-moving target, and each time through the submit-reject cycle takes long enough for the data to be outdated.
So much for the history. Now the question: how should I cite this paper? Specifically, what date should I give it? If I cite it as from 2004, it will give the misleading impression that the paper has been available for ten years; but if I cite it as from 2014, it will imply that it’s been worked on at some point in the last ten years. Both approaches seem misleading to me.
At the moment, I am citing it as “Taylor (2014 for 2004)”, which seems to more or less capture what’s meant, but I don’t know whether it’s an established convention. Is there an established convention?
Releated: where in mv publications list should it appear? At present I am sorting it under 2014, since that’s when it came out; but should it be under 2004, when it was written? I guess publication date is the one to go far — after all, it’s not unusual even now for papers to spend a year or more in press, and it’s the later (publication) date that’s cited.
Help me out. How should this be done?
References
- Taylor, Michael P. 2014 (written in 2004). A survey of dinosaur diversity by clade, age, place of discovery and year of description. PeerJ PrePrints 2:e434v1. doi: 10.7287/peerj.preprints.434v1
Sauropod Vertebra Picture of the Week 
July 29, 2014 at 11:35 am
Taylor manuscript 2004, released as peerJ preprint 2014. In text citation might be (Taylor 2004/2014)
July 29, 2014 at 3:12 pm
I am certainly not claiming this as the definitive answer, but shouldn’t you just cite it with the 2014 date?
Of course, all published articles were written before their date of actual publication and (depending on the speed of their journal) some of them may have been written years earlier . Those articles are always cited with the date of publication (not the date of original completion).
Therefore, I would say you would cite it with the clear date of public availability (i.e. the 2014 publication date) but perhaps explain somewhere in the text (but separate to the formal citation text) about the age issue.
July 29, 2014 at 6:24 pm
I’ve read somewhere a date like this: (2014 [2004]). I’m not sure where I saw it, but I think it was on something that was published, and republished by two sources. Again, not sure, so I can’t be any more specific.
July 29, 2014 at 7:35 pm
Have the same opinion as Peter Binfield, with a small added detail: what is the role of a citation? The only important info concerns the URI of the thing, so as long as you give the link to the PeerJ preprint, the rest is less relevant.
July 29, 2014 at 7:40 pm
Well, I don’t quite agree with Chorasimilarity there. The URI is fine as a way for a machine to understand what the cited item is, but it’s useless to a human. Whereas when someone who’s familiar with sauropods sees a citation of, say Wilson and Sereno (1998) or Hatcher (1901), they know immediately what papers are being referred to. That’s a valuable shorthand (and one reason that I detest numbered references).
July 29, 2014 at 7:59 pm
In the text could be something like (Taylor 2004/2014) for example. As you write in the post linked in your comment, really any other system which supposes a shorthand for the cite and a list of references at the end of the article is from another age. Another thing is that your preprint is a thing now, independent on the community common language of the moment.
July 29, 2014 at 8:02 pm
“Your preprint is a thing now, independent on the community common language of the moment.”
Right, exactly — the preprint is the thing of the Shiny Digital Future. We as a community have yet to figure out how to treat it. We don’t know how to cite it. We don’t know how much it “counts” in terms of reputation. We don’t know what old-style journals will and won’t allow papers to cite it.
We’re just figuring all this stuff out. We’re on the frontier. It’s pretty exciting.
July 29, 2014 at 10:42 pm
Alternatively, just explain, in the text, that the PeerJ preprint (Taylor 2014) was written in 2004 and never otherwise published. Trying to wrap up the story of the paper into a citation format and expecting people to understand it is a bit weird, now I think of it. Nothing wrong with a bit of personal backstory.
July 30, 2014 at 3:21 am
My guess: 2014, with a note in the text where it is referred to saying that it was written in 2004.
July 30, 2014 at 9:08 am
“Trying to wrap up the story of the paper into a citation format and expecting people to understand it is a bit weird, now I think of it.”
I don’t think it’s weird, I think it’s just unusual — so far. But I hope we’re going to see a lot of this kind of thing in the future: perfectly good science liberated from people’s personal hard drives and made available to the world years after it was initially written. So it will help us all if we have a simple, concise format for expressing that. If one doesn’t already exist, we should invent one.
July 30, 2014 at 1:01 pm
I would go with the citation that PeerJ suggest as to do otherwise could be confusing. Your Author Comment explains its history. However, if you’ve made it available earlier, eg through a website, then that would be an alternative citation.
July 30, 2014 at 1:15 pm
Anna makes a good point: the publication venue itself (in this case, PeerJ preprints) gives a suggested citation. It includes only the year 2014. So that can be construed as an argument that I should do the same. (On the other hand it could equally be an argument that PeerJ Preprints should allow authors to specify a date when the paper was written, and include in the citation. Especially if, as I hope, it becomes common practice to rescue old papers in this way.)
July 30, 2014 at 7:39 pm
I would be tempted to ask – what does your preferred style recommend you do if you cite a modern edition of a classic monograph; is it Jones (1858) – date of work – or Jones (1973) – date of physical copy?
It seems to be more or less the same issue. Ditto modern editions of otherwise unpublished papers – date of the text or date of the publication?
July 30, 2014 at 8:25 pm
Things may differ between fields, but from the palaeontologist’s perspective I don’t see that the reprinting date is of interest. I would unhesitatingly use 1858, and it wouldn’t even occur to me that there was a choice to make.
August 10, 2014 at 9:52 am
It’s a bit like, but also unlike, the situation of a thesis that was unpublished for some years, and later made available. The date and form in which a thesis is examined and/or accepted can reasonably be considered as creating a version of record (even if nobody else ever sees that precise version, e.g. if the later-posted version incorporates additional corrections), but having a submitted ms rejected without review probably doesn’t.
August 11, 2014 at 8:16 am
The thesis analogy is interesting. When I posted my own dissertation on arXiv six years after it was examined and accepted, it didn’t even occur to me to use the then-current date. It’s Taylor 2007, just as it had been for the previous six years if you wanted to cite it. That makes me think that perhaps the same is true in this case.