What is the scope for the contents in EarthArXiv
Bruce Caron
Public
Seen by 110
Question: Should EarthArXiv accept papers that are interesting for the Earth sciences but are non-standard in the sense that they may not be “preprints” as they are not really designed to be submitted later as science artifacts (although they might get published as a part of the editorial side of a science journal)? For example: aspirational essays on how changes to standard work flows might lead to new discoveries, critiques of how Earth data formats are sharable, theoretical essays on meta-science issues such as the impact of open science practices for Earth science.
EarthArXiv could be a great home for these. But does their inclusion affect how the archive is perceived as a place where “real science” is found?
Myself, I think the perception issue is misplaced. In my view, the service is designed to hold and expose (through search) all content that might be of value for the Earth (and Space) sciences. Nobody is reading the archive as a journal. Since we don’t have a paper product, we don’t need to be careful about the amount of content that gets included.
Perhaps EarthArXiv can be a leader in opening up the notion of “publication” as a first-step, and not a final step in research, and can be a home for content that might be excluded from existing forms of science journals.
What are your thoughts?
Matt Spitzer Sun 3 Jun 2018 8:24PM
Great discussion. Right now there is not a way to segment content within a single OSF-hosted preprint service to fulfill simultaneous use cases. Just to add that the use case described for figshare might be better served by submitting to the generalist repository at OSF Preprints at https://osf.io/preprints/submit. This is an un-moderated, all-discipline, repository that provides DOIs, citations etc.. The advantage being that if tagged, or submitted with an appropriate subject taxonomy, it would be discoverable alongside EarthArXiv content when searching at https://osf.io/preprints/discover
Christopher Jackson Sun 3 Jun 2018 8:38PM
He’s good. Real good...
Bruce Caron Sun 3 Jun 2018 8:55PM
I really like this conversation. And with Loomio, we can come back to it later too, or at least find it and remember it. Blogs seem like edge cases we don't need to go into now (figshare is an easy current solution). Better to look at capabilities that OSF already has that can add value when someone uploads a pre- post-print. This will also help us differentiate our preprint service from non-OSF ones. These options can be opt-in capabilities, and help support new cultural practices that are not tied to old paper journal ones. @christopherjackson1 ... I'm good with the current guidelines, the Advisory Council did a great job.
Eric-Geochem Wed 4 Mar 2020 4:04PM
It is very interesting that a recent editorial published by Science was echoed with the same mentality: "instead of weeding out, we should be weeding in.".
As a preprint repository, being open is not only about the inclusiveness of research content/ format but also the creativeness of scientific community.
In my mind, an acceptable submission for EarthArXiv would be legit, relevant to the geosciences borderline, and readable of decent clarity.
brandon Tue 8 Dec 2020 8:59PM
Apparently ESSOAr will start publishing peer reviewed Jupyter notebooks. On the face of it, I am interested.
I'm not really a 'keeping up with the Jones's' person, but it did get me thinking... what would the CDL framework need to publish non-PDF artifacts? Obviously, Jupyter notebooks could be one example, another could be RO-crates. What about a DOI which is a link to a manubot generated manuscript?
I find this interesting. Others? Thoughts?
Bruce Caron Wed 9 Dec 2020 3:43PM
One of the founding reasons for EarthArXiv was to open up new sharing capabilities not available from commercial publishers... so let's keep looking at where this can happen. (We are still getting the kinks out of a new platform, so nothing to do near-near term.
Daniel Nüst Thu 10 Dec 2020 9:57AM
Thanks @brandon for the pointer! I think moving away from the PDF is a core challenge of scholarly communication, and it would be great to see EarthArXiv take a lead here, or pilot something. Similarly to arXiv who partner with Papers With Code for machine learning preprints.
My spontaneous suggestion would be to archive Binder-ready repositories, so I can open a Jupyter Notebook or R Studio workflow based on an EarthArXiv DOI. An important advantage of this would be separation of expertise and resources, i.e., EarthArXiv probably does not want to get into the infrastructure game and provide computing power.
Evan Goldstein Thu 10 Dec 2020 2:46PM
Papers with Code just announced an expansion to other disciplines:
https://twitter.com/paperswithcode/status/1337044995686207490?s=21
brandon Wed 16 Dec 2020 8:18PM
Thanks for the links, @Daniel Nüst & @Evan Goldstein. Those are promising developments!
Christopher Jackson · Sun 3 Jun 2018 8:07PM
So, where have we got to with this? Personally, I like the suggestion that something like FigShare, personal blogs, etc, might be a more appropriate venue for this type of material, even if it then is not discoverable within the OSF search framework. I also agree with the sentiments above that somehow protecting the EA 'brand' might be valuable. @bcaron: seeing as you raised this issue, how do you feel now? And those in support of hosting this type of material, would you like to mount further arguments? It's an interesting this to discuss/think about...!