pitkant
commented
1 year ago Fixed in 7597740:
In the previous iteration the citation seemed to be a mishmash of several different citation styles:
(C) Leo Lahti, Janne Huovari, Markus Kainu, Przemyslaw Biecek. Retrieval and
analysis of Eurostat open data with the eurostat package. R Journal
9(1):385-392, 2017. doi: 10.32614/RJ-2017-019 Package URL:
http://ropengov.github.io/eurostat Article URL:
https://journal.r-project.org/archive/2017/RJ-2017-019/index.html- (C) is unnecessary here
- Year is after the Journal name, when it usually is after author names. Volume 9 number 1 is sufficient to recognise the release time of the journal.
- There is doi which is sufficient but also package URL and article URL which are redundant in a reference to a journal
R-exts states that "In case a bibentry contains LaTeX markup (e.g., for accented characters or mathematical symbols), it may be necessary to provide a text representation to be used for printing via the textVersion argument to bibentry" and bibentry help files that "Only if special LaTeX macros (e.g., math formatting) or special characters (e.g., with accents) are necessary, a textVersion should be provided.". Therefore it seems that textVersion is not needed but as users could just use print(citation, style = "text") but I think it's nice for end user convenience. However, we should be consistent.
The new textVersion citation is styled after Harvard:
Lahti L., Huovari J., Kainu M., and Biecek P. (2017). Retrieval and analysis of
Eurostat open data with the eurostat package. R Journal 9(1), pp. 385-392. doi:
10.32614/RJ-2017-019Why Harvard? Because it is supposedly used in almost all disciplines whereas other popular styles such as APA, MLA and Chicago are more used in specific disciplines
It is also incidentally quite close to what print(citation("eurostat"), style = "text") also outputs:
Lahti L, Huovari J, Kainu M, Biecek P (2017). “Retrieval and Analysis of Eurostat
Open Data with the eurostat Package.” _The R Journal_, *9*(1), 385-392.
doi:10.32614/RJ-2017-019 <https://doi.org/10.32614/RJ-2017-019>,
<https://doi.org/10.32614/RJ-2017-019>.rOpenSci states that their recommended way of styling a citation is
## To cite package 'magick' in publications use:
##
## Jeroen Ooms (2021). magick: Advanced Graphics and Image-Processing in
## R. R package version 2.7.3. https://CRAN.R-project.org/package=magick
##
## A BibTeX entry for LaTeX users is
##
## @Manual{,
## title = {magick: Advanced Graphics and Image-Processing in R},
## author = {Jeroen Ooms},
## year = {2021},
## note = {R package version 2.7.3},
## url = {https://CRAN.R-project.org/package=magick},
## }which is somewhat close to what the old solution was. I'm not entirely sure what citation format "Firstname Lastname (year). Title. Note. URL" conforms to so maybe it's best to use Harvard.
antagomir
Providing clear and concise way to cite the package is essential in encouraging users to cite the software they use. Traditionally, writing a software publication in a journal may have been seen as a more legitimate way to provide citable reference to academic users, but in recent years actors such as FORCE11 has encouraged to cite software directly. From Software citation principles:
The current citation for Eurostat package is:
The eurostat package has a published article in R Journal from 2017. Citing the article fulfils the 1. Importance requirement, as scholarly articles tend to be cited. In some ways 3. Unique identification, 4. Persistence and 5. Accessibility are also achieved by linking to the R Journal Article that in turn links to CRAN repository with a persistent URL. CRAN repository stores and archive of the package versions even if it was removed from CRAN. However, citing the article does not attribute contributions made to the package after 2017 (2. Credit and attribution), and it does not mention which version of the software was used in the analysis (6. Specificity).
Indeed, Katz et al paper Software Citation Implementation Challenges states explicitly that "R has guidance already, and that guidance does not match the software citation principles. For instance the guidance provided by the R Project does not include a version number or license information". R guidance seems to be mostly geared towards traditional academic journal and book citations, since the only example it offers is for a bibtype "book".
In aforementioned FORCE11 Software citation principles it is stated:
Therefore we should at least add a proper software citation to eurostat package in addition to retaining the R Journal citation.