[dinosaur] Years of publication, was Re: Sue the T-rex story and controversy over access to fossils (free pdf)

Pet peeve alert:

> Gesendet:=C2=A0Mittwoch, 08=2E Januar 2020 um 12:46 Uhr
> Von:=C2=A0"Thomas Richard Holtz" <tholtz@umd=2Eedu>
>
> Formal date for most journals is regarded as the publishing of the PRINT=
ED volume=2E Hence 2020=2E

"Formal" means what?

If you follow the link to the article, it says "Published: 31 December 201=
9", and then under the author's name it says "History and Philosophy of the=
 Life Sciences volume 42, Article number: 2 (2019)"=2E

Several big publishers indicate the date of online publication by "Publish=
ed:" or "Version of record online:"=2E

The International Code of Zoological Nomenclature recognizes the date of o=
nline publication as the date of publication if the paper has been register=
ed in ZooBank and contains evidence of that=2E Whether the paper is afterwa=
rds printed (two calendar years later as in the Journal of Systematic Palae=
ontology=2E=2E=2E) is irrelevant in that case=2E

Some journals publish accepted manuscripts online=2E Those differ from the=
 final published version in layout, including pagination, and at least pote=
ntially in content, because they haven't been proofread=2E These clearly sh=
ould not count for the purpose of establishing the date of publication=2E

But most journals that appear online and in print publish a version online=
 that differs from the later print version only in lacking permanent page a=
nd usually volume & issue numbers; the content is immutable, including whic=
h letters appear on which page=2E So why should we PRETEND that a publisher=
 can RETROACTIVELY change the date of publication of a paper by printing it=
?

In reality, how you have to cite a paper in your manuscript depends not on=
 the journal the paper appeared in, or on its publisher; it depends on the =
always secret house style of the journal you're submitting your manuscript =
to=2E PeerJ, for example, changed all publication dates to print dates in t=
he proofs of my huge paper=2E I tried to argue with them, they never answer=
ed, and the print dates are now published=2E

It's not going to matter in the present case where a paper was published o=
n the last day of the year and printed very early the next year=2E But in p=
rinciple, to cite a date that lies a year or two after the year of publicat=
ion is actively misleading about the contents: it makes papers appear more =
up-to-date than they are=2E

Also, I've seen "2017" papers that were responses to "2018" papers=2E Of c=
ourse they were published later than the papers they responded to; they wer=
e just printed faster=2E