medrxivr blogpost

[mcguinlu]

Good morning @ropensci/blog-editors,

Thanks for your patience, as I appreciate this is a day late! I had encountered issues when trying to previewing the post locally with Hugo, but I've checked it using the Netlify build and everything looks as it should.

As this is my second blogpost ever, very happy to incorporate any feedback you have!

Thanks for the opportunity to contribute to the blog,

Luke

[steffilazerte]

@mcguinlu No worries! I'll take a look at it today and get back to you soon. Thanks!

[steffilazerte]

Hi @mcguinlu

You said this is your second blog post (congrats)! I really love how awesome the medrxivr package is. It's always great to see a problem solved so elegantly. Your use cases are so lovely and clean and really show case how this package can help so many researchers. I do love my job as an editor :)

In general, most of my comments are stylistic, I think you may have missed some of the check boxes on the author submission list. But those are very quick fixes. The content is well organized, clear and easy to understand. And, as I mentioned above, the use cases are great.

The one thing I'm a bit uncomfortable with is some of the really frustrated tweets. Clearly the interface with medRxiv leaves something to be desired, but I don't want you or rOpenSci to look like you're just bashing medRxiv. I think you illustrate the drawbacks in the text clearly and with grace, without the need for the more negative tweets.

Thanks again for your well written post and useful package!

• post follows Content Guidelines
• post follows Style Guide
• title is in Title Case
• publication date is ok
• alternative text of images is informative
• Twitter metadata looks ok (paste post preview link in Twitter card validator; might have to click twice on "Preview card")
• author metadata is provided with correct folder name
• html not included in pull request of Rmd post
• I ran roblog::ro_lint_md() on index.md
• I ran roblog::ro_check_urls() on index.md
• I ran a spell-check on index.md
• YAML subject tags are ok
• YAML package-version included
• YAML subject tags - software peer review, packagename, R; add "community" for post by non-staff non-editor
• acknowledges and links to reviewers
• links to peer review thread

[steffilazerte]

Oh and if I could get your revisions by Friday or Monday, that'd be great!

[mcguinlu]

Hi @steffilazerte 👋

Thanks a million for your helpful comments - hopefully I have addressed them in my recent commits. I completely agree about the tweets - I just wanted to get across that it's not just me that has encountered these issues and that they are recognised problems in the community, but I can see in retrospect that they come across a bit strong! I've removed them now in favour of the text.

You are right to say I missed a few items on the checklist - in actual fact, I missed the checklist completely! 🤦‍♂️ Very sorry about that, but I think I have addressed all of the items on it now. Just to flag that roblog::ro_check_url() was not happy with some of the URLs in the post and I am not entirely sure why. . . . I've reproduced the output below, but just let me know if anything needs fixing:

Unknown option: verb
Unknown option: verb
Unknown option: verb
Unknown option: verb
Unknown option: verb
Unknown option: verb
Unknown option: verb
Unknown option: verb
Possibly broken URLs: https://www.medrxiv.org/, https://docs.ropensci.org/medrxivr/, https://www.biorxiv.org/, https://github.com/tts, https://github.com/njahn82, https://github.com/ropensci/software-review/issues/380, https://docs.ropensci.org/medrxivr/, https://doi.org/doi:10.1136/bmj.l2301.

[steffilazerte]

That's great! Thanks for your speedy edits. I'm a bit slow this week (got a cold, bah! With so much social distancing I thought that was impossible!) but I'll hopefully finish the review this afternoon.

[steffilazerte]

This looks great! I have one typo fix and two suggestions for the final code block (feel free to take or leave them).
Thanks for your 🚀 work

[steffilazerte]

I had no idea there was such thing as a NEAR operator. That is SO cool

[mcguinlu]

Yeah, you can currently use it in medrxivr via some pretty ugly regex syntax, but implementing it in a nicer fashion is definitely on the horizon!

[mcguinlu]

Great, thanks! Edits committed, plus one small final change from me (short comment in the code block just before the results object is printed).

Wishing you a very nice weekend, and a speedy recovery from the cold!

[steffilazerte]

Awesome, we're good to go! Thank you for your post, I'll publish and tweet on Tuesday 😸

[stefaniebutland]

@mcguinlu I just noticed that in the post, the package itself isn't mentioned until "Use cases". Would you consider adding a sentence or two right before Background so people know what the post and package are about? Maybe consider adding to the title? Searching medRxiv and bioRxiv Preprint Data with the medrxivr package.

cc @steffilazerte

[steffilazerte]

@mcguinlu That's right, I completely missed that you removed a paragraph from the beginning, definitely consider adding it back in, I quite liked it 😄

However, the native search functionality available on the medRxiv website is not suitable for
use in systematic reviews, due to a number of limitations. These limitations, presented below,
prompted the development of [medrxivr](https://docs.ropensci.org/medrxivr) R package which 
provides access to, and tools for searching, medRxiv preprint metadata. This post will also
summarize the key functionality of the package with respect to two key use cases. Note that 
medrxivr allows users to access and search bioRxiv as well as medRxiv preprint metadata, 
as both repositories are run by the Cold Spring Harbor Laboratory and so have a similar 
API endpoint/native website search functionality, but all illustrating examples in this post are
taken from medRxiv.

[mcguinlu]

Whoops - not entirely sure how I managed to delete that paragraph, but have re-added it now! Great catch @stefaniebutland!