Thursday, May 16, 2019 From rOpenSci (https://ropensci.org/blog/2019/05/16/dev-guide-update/). Except where otherwise noted, content on this site is licensed under the CC-BY license.
As announced in our recent post about updates to our Software Peer Review system, all our package development, review and maintenance is available as an online book. Our goal is to update it approximately quarterly so it’s already time to present its second official version! You can read the changelog or this blog post to find out what’s new in our dev guide 0.2.0!
Let’s start with very exciting news, the dev guide now has a cover, designed by Oz Locke from Locke Creatives!
It shows a package production line with small humans discussing, examining and promoting packages, all of this in white on rOpenSci blue. Speaking of rOpenSci branding, thanks to Will Landau’s help, the online book now features the rOpenSci hex logo as favicon.
Two changes improved the accessibility of the book, thanks to feedback from readers. First, the book became more legible by the use of a darker blue for links. Thanks a lot to Kevin Wright for telling us how hard it was to read the links in the former color! Then, the README of the GitHub repository holding the source of our blog got more accessible thanks to a contribution by Katrin Leinweber. In particular, Katrin made us aware of guidance about why not to use “click here” for links.
Thanks to a question by Robert M Flight, our style guidance now states that it is fine to use either an arrow or an equal sign for assignment as long as the choice is consistent within the whole package. You can choose to use
<- as long you are consistent with one choice within your package. We recommend avoiding the use of
-> for assignment within a package. If you do use
<- throughout your package, and you also use
R6 in that package, you’ll be forced to use
= for assignment within your
R6Class construction - this is not considered inconsistency beause you can’t use
<- in this case.
We added a new continuous integration requirement: package maintainers must now use
oldrel R versions on Travis, not only R-release.
In the section about recommended scaffolding we already recommended xml2 over XML for most cases but now we’ve added a link to Daniel Nüst’s very neat notes about migration from XML to xml2.
Three changes relate to what happens after a package has been accepted.
The review template now includes a question asking the reviewer for consent to be added to DESCRIPTION in review template, should the author(s) deem it appropriate. For more context, refer to [our blog post “Thanking Your Reviewers: Gratitude through Semantic Metadata”] (/blog/2018/03/16/thanking-reviewers-in-metadata/).
rOpenSci community manager Stefanie Butland and other authors recently reported how our software review guidelines ended up being used “in the wild” for a scientific paper. Knowing about such use cases makes us happy and helps us assess the usefulness of our material beyond our own system, so we’ve now added the following wish to several places in our guide (intro to our software review system, reviewer guide, chapter about contributing to rOpenSci):
If you use our standards/checklists etc. when reviewing software elsewhere, do tell the recipients (e.g. journal editors, students, internal code review) that they came from rOpenSci, and tell us in our public forum, or privately by email.
The contributing guide now contains more reasons to contribute, and the approval template for editors now features more specific guidance about writing a blog post or tech note about an approved package.
In this post we summarized the changes incorporated into our online book “rOpenSci Packages: Development, Maintenance, and Peer Review” over the last three months. We are very grateful for all contributions that made this release possible. Now, if you have any feedback about the book, you too can head to the issue tracker!