“In the past, PLoS has gone to great lengths to ensure that taxanomic papers published in PLoS One meet the ICZN standards. Last year PLoS One published a paper describing a revision of several ant genera.
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0001787
Cognizant of the ICZN standards, PLoS One ensured that print copies of the article were deposited in appropriate libraries. This was described explicitly in the paper.”

However, from an ICZN-nomenclatural perspective this is quite ambivalent. The actual ant-publication says: “In accordance with section 8.6 of the ICZN’s International Code of Zoological Nomenclature, we have deposited copies of this article at the following five publicly accessible libraries: [...]” The statement in the article could be quite allright, if (and only if) these copies are CD’s, DVD’s, etc. However, if these copies are paper printouts, it is nonsense, as Art. 8.6 only deals with “[...] a work produced after 1999 by a method other than printing on paper”. On the face of it, judging only by the PLos ONE-paper, it looks in order, as it is not made clear that the “copies” are not the required CD’s, DVD’s, etc, so a reviewer might not catch this, nor would a reader, as he would suppose that the names were published on the 5 CD’s, DVD’s, etc that were deposited. However, apparently the reader would be wrong in his assumption: apparently these ant names were published in the paper printouts sent to the libraries, and the number of five libraries is an arbitrary choice, where the Code requires “numerous identical and durable copies”, that were “simultaneously obtainable”. Anyway, there is nothing “explicit” about it; the reader has to enquire further to find out what exactly did happen and how and where these names were actually published.

Note that if these ants were plants (with botanical names), this procedure of sending out paper printouts to “botanical institutions with libraries accessible to botanists” would work, while sending out CD’s, DVD’s, etc would not. Here too, it would not be required to specify this in the paper itself.

As what would work for a zoological name would not necessarily work for a botanical name (and vice versa), PLoS ONE’s adopting a policy that meets with the requirements of the animal Code (the ICZN) is not guaranteed to be sufficient for the future. Nor is it guaranteed that something that will work now will keep working in the future. These rulebooks are adjusted from time to time, sometimes because of incidents like this one.

(BTW: The one point that I really disagree on with Richard Pyle was his suggestion to alter the pdf to change the date. I am much happier with the solution actually adopted, namely to add an explanatory footer.)

The pending proposed Amendment to the ICZN Code for allowing electronic forms of publication (see: http://www.iczn.org/electronic_publication.html) is currently in review, as is required for all such major amendments to the Code. This process will likely be completed within the next year, and if adopted, the amendment should go into effect at that time.

What will require “a few years” to be published is the next (Fifth) Edition of the ICZN Code (see: http://iczn.ansp.org/). Presumably, this Edition of the Code will also support the electronic publication of nomenclatural acts (especially if the proposed amendment to the existing 4th Edition of the Code is approved).

A print-run of fifty copies of the paper has been created on May 21st. The top sheet of each copy has the following text appended to the footer: “This document was produced by a method that assures numerous identical & durable copies, and those copies were simultaneously obtainable for the purpose of providing a public and permanent scientific record, in accordance with Article 8.1 of the International Code of Zoological Nomenclature. Date of publication: 21st May 2009”

Apart from this wording, these copies are identical to the electronic version that is freely available from our web site at: http://www.plosone.org/article/fetchObjectAttachment.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0005723&representation=PDF

These copies are now obtainable from our offices at 185 Berry Street, Suite 3100, San Francisco, CA 94107, USA. Anyone who requests a copy, and tenders a fee of $10 (towards the cost of postage and printing) will receive a copy.

Having made the printed copies available, we have been told by the individuals named above that we have conformed with the relevant ICZN codes. They have also indicated that the proposed resolution is an interim step, which should meet the requirements of the Code until a formal amendment is published within the next few years.

We are very grateful to the ICZN for their actions to resolve this matter.

and none of us really wants this thing to be called Rioarribasimia.

I’m still ROTFL about this, but it’s not actually funny at all…

And it gets still worse. Doesn’t everything that applies to Darwinius also apply to Maiacetus?

This is fishy. Stop making excuses and start convincing us, the scientific community, that:

A) Your editorial board is going to develop policy that will prevent such things from happening again

B) Your editorial board is going to consider your core principles of “Excellence” and “Scientific Integrity” to be of equal importance as “Open Access” and “Science as a Public Resource.”

C) Your editorial board recognizes that mistakes of this caliber reflect not only on the authors of the paper, but also on the editorial practices of the journal. This is not simply an authorial mistake; it is an editorial mistake as well.

D) Scientists who choose to publish their work in PLoS will get the editorial attention that their work deserves. Our research and our ideas reflect on you, and your journal reflects on us. As scientists, we have a responsibility to ensure that our work is as rigorous as possible, as accurate as possible, and as clearly communicated as possible. As editors, it is your responsibility to ensure that our work is clear, is rigorous, and adheres to the standards of our community. When these responsibilities are met, we both profit from it; your journal gains prestige and our work helps us contribute to our community. When they are not met, both the journal and the authors look sloppy and amateurish.

1) A new species description published in Nature, for which a PDF copy is distributed online weeks prior to the date when the paper-printed copies are obtainable.

2) A new species description published in Zootaxa, for which a PDF copy is distributed concurrently with the date when the paper-printed issues are obtainable.

3) A new species description published in PLoS, originally with the intent of existing only in its electronic form(s), but then printed as numerous identical copies a week

later, which are made obtainable free of charge or by purchase.

4) A new species description created as an MS Word file and sent around to the author’s colleagues for informal review, which is later printed as numerous identical copies on

the author’s laser printer in his home office, which he intends for the permanent scientific record and is perfectly willing to mail out copies to anyone who asks for it.

From the perspective of a scientist, these may all seem very different.

From the perspective of the Code, they are essentially indistinguishable. In all cases, the electronic version of the publication is irrelevant. It matters not whether the

electronic version is identical to the paper-printed version, slightly changed, or completely different. Nor does it matter whether the electronic version was distributed

prior to, or after the paper-based version. From the perspective of the Code, it simply does not exist. All that matters (in the context of the current conversation) is

whether a document was issued for the purpose of providing a public and permanent scientific record, was produced in an edition containing simultaneously obtainable copies by a

method that assures numerous identical and durable copies, and that such copies are obtainable, when first issued, free of charge or by purchase. The date of availability is

the date on which the numerous identical paper copies were simultaneously obtainable.

The alternatives are to believe that adherence to taxonomic standards and/or good peer-review and editorial practice are not priorities for the PLoS staff, which are not good alternatives.

First of all, the comments about Article 8.6 (and associated requirements to distribute copies to libraries, mentions of such within the publication, etc.) is misdirected in this case. Art. 8.6 applies only to publications on “durable” media other than paper. It was intended to cover things such as CD-ROMs and the like. Although PLoSONE attempted to conform to art 8.6 on a previous article describing new species (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0001787), no such attempt was made in the case of Darwinius, because no “durable” media was involved. Thus, for the time being, we can ignore the implications of Art. 8.6.

The critical Code Article in this case is 8.1:

8.1. Criteria to be met. A work must satisfy the following criteria:

8.1.1. it must be issued for the purpose of providing a public and permanent scientific record,

8.1.2. it must be obtainable, when first issued, free of charge or by purchase, and

8.1.3. it must have been produced in an edition containing simultaneously obtainable copies by a method that assures numerous identical and durable copies.

8.1.1 is probably safe to assume in the case of the PLoS article. And it probably excludes all the paper-printed media reports on the issue (as does Article 16.1).

8.1.2 and 8.1.3 are the ones that we should be focusing on. Specifically:

1) Did PLoS create “numerous” (technically >1) paper copies of the article (as they did for the ant article linked above)?

2) If so, were these paper copies “simultaneously obtainable”?

If the answer to both of these questions is “Yes”, then the name is available from those simultaneously obtainable paper copies (not from the online edition). As it happens, there is no requirement to distribute those paper copies to any libraries (or anyone else, for that matter) — they need only be “obtainable”. (The “5 major publicly accessible libraries” thing is only relevant to durable copies produced on something other than paper as per Art. 8.6, which as I said doesn’t apply in this case.)

However, if the answer to either of the above questions is “no” (and I believe I have confirmation now that this is the case), then the name is not currently available. It could be made available, at any time, by anyone, simply by producing “numerous” paper copies of a description (that also fulfills all the other requirements of the Code) and making those copies simultaneously obtainable. This is where things get a bit dangerous for the authors of the PLoS article, in the sense that someone could “scoop” them on the name.

The good news is that this doesn’t seem to have happened yet. As stated above, we can safely eliminate all the newspaper articles and such — even if they were printed on paper. The real danger, however, is that someone might intentionally and maliciously pre-empt the original authors by establishing their own name (or the same name — doesn’t matter) based on the same type specimen. This would be Michael Taylor’s option 3, and I agree with him that this would be the WORST possible outcome — particularly for the perpetrators of such unethical behavior themselves (assuming they value their reputations and careers).

In my mind, the solution is striaghtforward: PLoS should print “numerous” paper copies of the original description, and then make them simultaneously obtainable. In doing so, they would fulfill the requirements of the Code, and the names contained therein would become available as of the date on which the numerous paper copies were obtainable. In the ideal case, I think the paper copy should be an exact representation of the PDF, with one exception: there ought to be a small change: the date following the word “Published” on page 1 should be the date on which the paper copies were made obtainable; not the date on which the PDF was released. This last bit is certainly not rquired for Code compliance, but it would help mitigate future confusion by allowing the distinction between the two versions (PDF and paper), and stating the correct date of availablity (which is necessary for purposes of establishing priority under the Code).

In the near-term future, I think all of this fuss and confusion can be reduced or eliminated if journals that produce electronic publications that contain Code-goverend nomenclatural acts simply created “numerous” (technically 2 or more, but 50 is probably a better number) paper copies, and made those copies obtainable free of charge or by purchase on the same date that the PDF is released.

It is really weird that PlosONE did not give any thought to establishing anything at a nomenclatural level, considering the size of the hype they organized. It is not a matter of overlooking a detail, but of just not caring at all.

The reasoning involving Art 8.6 would hold up if there was a simultaneously produced CD or DVD. It would have been easy to provide (and mail out) CD’s, DVD’s or printouts, but instead there apparently never was anything at all. I apologize for fixing on an (irrelevant) detail and overlooking the obvious. I should know better.

What Art. 8.6 explains, is how a “work produced by a method that does not employ printing on paper” can become published in the sense of the Code. But Art. 9.8 also explicitly excludes “text or illustrations distributed by means of electronic signals (e.g. by means of the World Wide Web)” from what can be (thus become) published work.

Art. 8.6 is entirely irrelevant to web-only journals as such. It applies only to works that are not printed on paper, but are nevertheless produced /physically/, using another method – e.g., as CD-ROMs. A web-based production can simply /never/ become published under the current Code, whatever you include in the text.

For the content of a web-only journal to become published under the Code, you must run a physical production in parallel to the web production. It is this physical production, and only it, that can become “published” – the web-based production itself remains technically “unpublished”. If this is done using a method that does not employ printed paper (e.g., the pdf on a series of identical CDs), then this physical production must satisfy 8.6. (But note that the required statement, as long as it is on the CD, could be in a separate file, outside the main pdf.) If this is done on printed paper, then 8.6 does not apply at all.

http://www.nlm.nih.gov/pubs/factsheets/errata.html#corrected

I would think this article qualifies. So, indeed all may be okay if we go with this solution after all.

Other solutions are:
1 the solution provided by Ellinor Michel: Print out the PlosONE Article in a way so that the print-out fulfills the requirements for publication. In that case the print-out is the work that counts, and that establishes the date of publication.
2. publish a new electronic paper that does conform to the requirements and that does not repeat the entire publication but only the nomenclaturally relevant material.