Frequently Asked Questions

General

What is the difference between the main repository and the sandbox?

The sandbox is meant to practice with the concepts and the procedures of the "Dataverse" software. The Datasets created on the sandbox are regularly deleted. A Dataset published on the sandbox will be visible by anyone visiting the website (until it is deleted), but it will not be indexed by search engines (e.g., by Google).
The main repository is the official repository for storing and preserving the research and observation data produced at IPGP. A Dataset created on the main repository cannot be deleted after its publication, and every modification will generate a new version (see the Procedures below).

I have an account on the sandbox, but it doesn't work for the main repository (or vice versa).

Accounts are not shared between the main repository and the sandbox. You have to create two separate accounts.

Where should I put my data? I'm confused about "Collections".

Datasets are organized into Collections.
- If you are experimenting in the sandbox, just use the main collection of the sandbox.
- If you need to upload data to the main repository, the choice of the Collection depends on the kind of data:
  - A Dataset associated with a peer-reviewed article should go into the main collection.
  - A Dataset associated with a funded research project should go into the project's collection.
  - A Dataset containing an observatory product should go into the observatory's collection.
- If you need to create a new Collection (e.g., for your funded research project), please contact rc-support@ipgp.fr.

I created an account, but I cannot see the button "Add Data".

Every account needs to be authorized to upload data (note, this is different from the "account activation" mail you received). To authorize your account, you must go to the main page of the Collection where you want to upload your data, then click on the "Contact" button on the upper right and write a message to introduce yourself and explain which kind of data you need to upload.

How can I migrate a Dataset from the sandbox to the main repository?

There is no automated migration tool. A solution is to copy the metadata fields from the sandbox and paste them into the main repository; then, you will need to re-upload the files to the main repository.

What's the difference between the DOI of my article and the DOI of the Dataset associated with my article?

Every digital object on the web can be identified by a DOI (Digital Object Identifier).
Your Dataset is a different object than your article: that's why it must have a different DOI.
It's important to create a link between your article and your Dataset. For that:
- Cite your Dataset in your article
- Add a reference to your article in the Dataset metadata.
See the "Procedures" below to learn how to perform the above steps.

File upload

Is there a limit on the number and size of files I can upload to a given Dataset?

The maximum number of files you can upload to a given Dataset is 1000 (one thousand). The maximum size of a file is 5 Gb. If you have more than 1000 files to upload, you can create one or more zip archives containing your files (each zip archive cannot exceed the maximum allowed size of 5 GB).

How can I keep my files organized?

There are different approaches (not mutually exclusive) to keep files organized:

Prepare a zip file on your computer containing your files organized into a directory structure. The zip file will be decompressed after uploading and the directory structure will be kept (click on the "Tree" view to show the directory structure). Note that if the zip file contains more than 1000 files, it will not be decompressed.
Use File Tags.
Add a "README" file to explain how your files are organized. Note: you might want this file to be the first of the list; for that, name it 00_README.txt.

I'm trying to upload a zip file (or any other archive), but it gets uncompressed. How to keep the file in zip format?

Every zip file (or any other archive) containing less than 1000 files will be uncompressed. If you need to keep the file in zip format, upload a zip file containing your zip file (double zip!): only the outermost zip file will be uncompressed 😉.

I uploaded an Excel or a CSV file and I got an "Ingest error".

Files containing tabular data (e.g., Excel, CSV) are "ingested" upon upload. The goal of the ingest process is to extract the data content from the file and archive it in an application-neutral, easily readable format. A file correctly ingested will get the .tab extension.
The ingest process might fail if the file is not correctly formatted (e.g., a row having less columns than the others). Try to:
Double-check your file for possible inconsistencies.
See whether it is possible to convert your Excel file to a simpler CSV format; then try again uploading it.

No preview is shown for my text file.

Text files need to have the .txt extension in order to be previewed.

Metadata

How to add another author?

To add multiple authors, click on the "+" sign on the right of the "Author" form.
Note that author names must be entered as "FamilyName, GivenName" (e.g., "Skywalker, Luke").

Where can I specify the geographic extent of my Dataset?

When you first create a Dataset, you are presented with a basic set of metadata fields. Enter the basic metadata, then click on "Save Dataset". You can now click on "Edit Dataset → Metadata" to add more metadata.
To specify the geographic extent, scroll to the bottom of the page and expand the "Geospatial Metadata" section.

Where can I specify the timespan of my Dataset?

When you first create a Dataset, you are presented with a basic set of metadata fields. Enter the basic metadata, then click on "Save Dataset". You can now click on "Edit Dataset → Metadata" to add more metadata.
To specify the timespan of your Dataset, use the field "Time Period Covered".

Where should I put the name of the article associated with my Dataset?

To specify the name of the article associated with your Dataset, use the field "Related Publication". Please remember to also specify the article's DOI by selecting "DOI" for "ID Type" and entering the article's DOI in the field "ID Number".

How to enter multiple keywords?

Do not separate keywords by a comma.
Enter multiple keywords by using the "+" sign on the right of the "Keyword" form.

I don't know which license to choose.

The choice of a Dataset's license terms depends on the status of your data:

If your data has restricted access, you should select "Custom Dataset Terms" and fill up the "Terms of Use" field to explain why your data is restricted (e.g., under embargo, sensitive data, etc.) and which are the conditions and procedures to request access to the Dataset.
If your data has open access, we advise using "Licence Ouverte / Open Licence 2.0", which is the official French license for open data. We also provide, for compatibility reasons, the "Creative Commons Attribution 4.0 International" (CC BY 4.0) license, as well as the "Open Data Commons Open Database License (ODbL) 1.0".
If you need to use a different open access data license, please select "Custom Dataset Terms" and fill up the "Terms of Use" with the information on the license.

Procedures

I'm submitting a manuscript to a journal and I want to make a Dataset available to the reviewers.

Please follow the steps below:

Create your Dataset, provide as much as possible information in the metadata, upload the files and choose a proper license.
Do not publish your Dataset! An unpublish Dataset has an inactive DOI which you can indicate in the manuscript. The DOI will be activated when the Dataset will be published, upon acceptance of the manuscript (see below).
Create a private sharing URL by clicking on "Edit Dataset → Private URL".
In the "Data Availability Statement" of your manuscript, write a paragraph similar to the following (using the actual DOI and private URL for your Dataset):

The data used in this manuscript is available through IPGP Research Collection (research-collection.ipgp.fr) at the following DOI: 10.18715/IPGP.2023.xxxxxxxx. Note that this DOI will be activated upon acceptance of the manuscript. In the meanwhile, the journal editors and the manuscript reviewers can privately and anonymously access the data through the following URL: https://dataverse.ipgp.fr/privateurl.xhtml?token=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx

My manuscript has been accepted, what should I do with the associated Dataset?

First of all, congratulations! 😄
Please follow the steps below:
1. The editor should ask you for a final version of the manuscript ("accepted version"), which will go to copy editing and, in some cases, published online while waiting for the copy edited version. In this version, remove the phrase on the private URL (see previous question).
2. As soon as the journal provides you with a DOI for the accepted article, enter this information into the "Related Publication" field of the Dataset metadata (see "Metadata" above).
3. Once your metadata is completed with the "Related Publication", you can publish your Dataset ("Publish Dataset" button on the upper right). The Dataset's DOI will become active. Note that, after publication, any further modification to the Dataset will generate a new version (see below).

My research project has been funded, and I would like to create a Collection to store the project's products.

Nice! Please contact rc-support@ipgp.fr to get instructions.

I would like to modify a published Dataset.

It is possible to modify the metadata and/or add/remove/modify one or more files of a published Dataset.
Please note that any modification to a published Dataset (even correcting a small typo) will require creating a new version of the Dataset. The previous versions will be kept for reference.

I would like to delete my published Dataset.

This is generally not possible! All Datasets published on the main repository are assigned a valid DOI and cannot be deleted. Please consider instead creating a new version of the Dataset (see the previous question). If you have a serious reason to delete a Dataset from the main repository, please contact rc-support@ipgp.fr.
If your question concerns the sandbox, please note that Datasets (published and unpublished) are periodically removed and the associated DOIs are not valid. If you have a serious reason to immediately delete a Dataset from the sandbox, please contact rc-support@ipgp.fr.