When adding a dataset it is missing a form field to add an external data repository web address
Miguel Silva
I've created a dataverse to host files and documents for each of the reviews I made to a Journal. Nowadays it is pretty much mandatory for authors to provide at least one data repository where it is possible to review and evaluate any supplementary data including experimental data. Anyone can evaluate more on this at Harvard's Dataverse here:
https://doi.org/10.7910/DVN/0AXHEA
That said, when creating a new dataset for this purpose it is missing a form field where I, as a reviewer, can input the web address of the data repository authors provided for me when reviewing a document for publication in a journal.
allow me to elaborate,
as a reviewer, I'm looking to help authors get published and at the same time I need to receive "credits" / reputation for each individual review I make and deliver. Moreover, with the current state of software-assisted tools and software automation, it is urgent to advance and include a data repository when submitting a document for publication.
(...)
I could go further here, about the importance of data repositories, however, I will skip it for now...
(...)
The role of a reviewer needs to go beyond the template, design, and layout evaluation of a scientific document, it needs to go beyond grammar and misspell correction, those tasks are more suited to a professional writer and a linguist with their software-assisted tools.
As a reviewer, I need to focus on what has real value in the submitted document: the proposed mathematics and models proposed for experimental data analysis, proposed methods utilized on both experimental campaigns and later when analyzing experimental data. This means a data repository is now mandatory, to me at least, so I as a reviewer can evaluate all previously stated on my own computer, alternatively on a virtual computer made available by the journal.
(...)
In the end, as a reviewer, I want to deliver a review where I am able to showcase the review made and also how it was made. For instance,
- a video recording of me testing the Python code provided by the authors showing it is actually working and outputting real results.
- I may need to add an Excel file with my own analysis and evaluation of the mathematical models used.
- a Jupiter notebook with my own custom code when performing some type of XAI (explainable AI)
- etc ...
To me all these files a reviewer produces need to be parked in a data repository. And linked later when the publication is accepted to be published in the journal.
---------------------------------------------------------------------------------
This conversation is also a feature request on GitHub
https://github.com/IQSS/dataverse/issues/10622
sbar...@g.harvard.edu
Miguel, Just resending my message that this was done in error. You obviously have a collection of "data." Please revive your deaccessioned deposit and if you need assistance with this, let me know.
Regards
Miguel Silva
Thank you for the replay.
and I totally agree that a dataset cannot change its status to "unaccession" like it happened this past time.
🫨
unfortunately, as in any real-time environment, I had to create a new dataset with a new DOI so I could send my review to the Journal.
I even added a bogus command delimited .csv file just to be sure it does not happen again.
On my end, problem solved, and "crisis" avoided . That;'s what matters at the end of the day.
👍
Miguel Tomas
Miguel Silva
😁