Database Conditions

Information about any conditions on access, deposition, sustainability or preservation

This portion of a database record provides metadata that the research community has determined to be highly relevant to their decision-making when choosing a repository. Therefore, by curating this section, you are increasing the likelihood that your resource will be discovered by prospective users.

Please note that the pulldowns for each of these fields are mandatory. The note and URL fields remain optional, where present.

The fields describing the Database conditions within the "Additional Information" tab are applicable to our database registry only

not found is an option available within all pulldown menus in this section. It should only be used when the information required to make an expert judgement about that particular field is simply not findable. This value should only be used when there is no other choice, as it tells the FAIRsharing community that the database being described has not provided this information at all.

Alignment with Existing Efforts

As is the case with all of our metadata, the database properties listed here have been created in response to the needs of our user community. Where we have explicit alignment of the community resources below, such alignment is provided in the tables associated with each field.

Alignment with RDA

FAIRsharing metadata is informed via collaborations with the RDA and its past outputs relating to database attributes. FAIRsharing is also represented on the RDA Data Repository Attributes WG, and implements the RDA recommendations that have resulted from this WG. The current version used in this alignment with the RDA DRA WG's RDA Common Descriptive Attributes of Research Data Repositories is 0.9. Where such an alignment exists within our database conditions, they are listed within the alignment table for that field. Otherwise, they are listed in the general alignment table directly below.

RDA DRA WG	FAIRsharing	Level	Comment
1 - Repository Name	Record Name	exact
2 - URL	Homepage	exact
3 - Country	Countries	exact
4 - Language	n/a	implied	FAIRsharing is an English-language resource, and by default the expectation is that, for every resource described in FAIRsharing, sufficient information is available in English to make an informed decision regarding it. Other languages may be described in the description field, but this is optional.
5 - Organisation	Organisations and Grants	exact
6 - Contact	Contact Information, Support Links	exact
7 - Description	Description	exact
8 - Research Area	Research subjects	exact	Uses the Subject Ontology.
9 - Persistent Identifiers	Relationships to Identifier schema records	exact	If a repository has a relationship implements or related to with an identifier schema record, then it is using that identifier. We are currently implementing a tag for those identifier records that are defined as PIDs by EOSC will be explictly tagged to aid decision making by our user community.
10 - Machine interoperability	Data Processes	exact
11 - Metadata	Relationships to model/format records	exact
12 - Curation, 13 - Terms of Deposit, 14 - Terms of Access	see below	see below
15 - Dataset Use License	Licences	exact
16 - Certification	Certification and Community Badges	exact
17 - Preservation	see below	see below

Alignment with NIH

These fields, where possible, also align with certain fields used within the NIH-Supported Data Sharing Resources (BMIC). Such alignments are listed within the alignment table for that field.

Data access condition

This item concerns the way in which the repository owners define access to their repository. What is the process through which access can be requested (and granted)? Is the data is freely available or subject to a request and approval process?

A resource is open when there are no restrictions on accessing its data, e.g. free registration or free accessibility of data. A resource is partially open when access to a subset of its data is restricted, e.g. paywall, for ethical or security considerations, or data protection issues. A resource is controlled when the entirety of its data has these restrictions. Finally, use not found if the database does not provide this information at all.

You can also optionally provide the URL of a document or webpage containing the data access conditions as well as a free-text notes field that provides a short summary.

Please note that the resource team (e.g. the developers of the database) define how accessible the data is. Even if users are allowed to hide their records, for example only giving access for pre-publication review, this does not affect the value of this field.

Data Access Condition is about whether or not there are restrictions imposed upon the access to the data within the database, as defined by the database developers. This is a completely separate topic to how the user may be allowed to use the data; data usage is covered by the licencing of the resource.

Community Effort	Attribute Name	Level	Comments
DRA WG Attributes	14 - Terms of Access	exact
NIH-Supported Data Sharing Resources (BMIC)	Data Access Policy (open / not open)	low	The primary determinant is cost-free re-use, even if there are restrictions on how that data is retrieved. This contrasts with FAIRsharing, where the primary determinant is the proportion of restricted data.

Data curation

This item is concerned with the review and annotation of the data performed by the repository (e.g. via a data submission tool that enforces some curation, automated, or manual curation). Does the repository curate its holdings? If so, choose from among Manual (where review and annotation of the data is performed by the submitter or a curation team), Automated (where curation is performed via programmatic methods), or both. Otherwise, select None. Finally, use not found if the database does not provide this information at all.

You can also optionally provide the URL of a document or webpage containing the data curation methodology as well as a free-text notes field that provides a short summary.

If there is no information regarding the database's curation efforts, you should leave this blank; this is to contrast with None, which should be used when there is clear information provided that the database does nothing to the data upon entry into the resource.

As an example, if a database pulls information from multiple sources, but states that it does no extra curation (e.g. if it is a data portal or a federated data source), then it is appropriate to use None. If, however, there is no explicit statement about curation after the data is added to the resource, you should use not found.

Community Effort	Attribute Name	Level	Comments
DRA WG Attributes	12 - Curation	exact

Data deposition condition

Deposition of data: are there any restrictions (e.g. by location, country, organization, etc.) or can anyone from anywhere deposit data? Data deposition is open when there are no restrictions on submitting data. Otherwise, data submission is controlled.

not applicable should be used whenever data deposition is not in scope for that database. Some examples include: a database that only stores data from a particular grant or study; a knowledgebase that ONLY pulls information from primary databases. Finally, use not found if the database does not provide this information at all.

You can also optionally provide the URL of a document or webpage for the the data deposition conditions as well as a free-text notes field that provides a short summary.

Community Effort	Attribute Name	Level	Comments
DRA WG Attributes	13 - Terms of Deposit	exact
NIH-Supported Data Sharing Resources (BMIC)	Data Deposit Timeframe (open / not open)	medium	if both data deposit timeframe and data submission policy are open, then FAIRsharing's data deposition condition should also be open, and vice versa. not open may be directly converted to FAIRsharing's controlled value, however this conversion must be checked if converting controlled to data deposit timeframe, as the reason why FAIRsharing may have selected controlled could be unrelated to the timeframe for deposition. This is because, while FAIRsharing would also use controlled when there is a time restriction on deposition, our primary determinant is whether any researcher with in-scope data is allowed to submit data.
NIH-Supported Data Sharing Resources (BMIC)	Data Submission Policy (open / not open)	medium	open in FAIRsharing would equate to open in this resource, however, in some cases an open data submission policy would result in a controlled value within FAIRsharing, e.g. BMIC will also provide an open value when only a particular set(s) of investigators may submit. not open can be directly converted to FAIRsharing's controlled value, and vice versa.

Data preservation policy

This item is concerned with the policy that details how the preservation of the data is ensured. Does the repository provide information on its data preservation policies?

You should provide the URL of a document or webpage containing the resource's data preservation policies as well as a free-text name field where you can provide a name or a short summary.

Community Effort	Attribute Name	Level	Comment
DRA WG Attributes	17 - Preservation	exact

Resource sustainability

This is where you can link to the document that gives information about sustainability plans for the repository if your database has a webpage or document that describes them.

You should provide the URL of a document or webpage containing the resource's sustainability plans as well as a free-text name field where you can provide a name or a short summary.

Community Effort	Attribute Name	Level	Comments
NIH-Supported Data Sharing Resources (BMIC)	Support funding duration (sustained / not sustained)	low	FAIRsharing provides a link to information about sustainability plans to let users know if such plans are in place. However, it makes no structured comment about the type of sustainability. However, the type of sustainability may optionally be provided in the free-text notes that are a part of this field.

Citation to related publications

Does the repository have a particular, standardized mechanism to link datasets to related articles or pre-prints? Answer yes or no; only use not found if the database does not provide this information at all.

Data access for pre-publication review

Does the repository have a mechanism to facilitate peer review of embargoed data? Answer yes or no; only use not found if the database does not provide this information at all.

Data Contact Information

Does the repository show data depositor contact information on dataset landing pages? The contact information for a piece of data within a database may be for an organisation or an individual (or group of individuals), depending on who has ownership of that data.

Answer yes or no; only use not found if the database does not provide this information at all.

For data contact information to be yes, we require an email address or other information (e.g. a link to a contact form for the owner of the data) that will allow the user of this repository to immediately contact the owner of the data. A name is not enough, nor is an ORCID. While an ORCID (usually) uniquely identifies a researcher, it does not require that an email address be visible on the researcher's ORCID profile. Names on their own are similarly problematic, as are publications.

If this field is set to yes, the expectation is that a user visiting this database will be able to easily contact the owner of data within the database without having to google or otherwise search for a valid method of contact.

Data versioning

Does the resource enable modifications to published data (e.g., to correct it or append additional information) and is there a process to distinguish, link and access all public versions of the data? Answer yes or no; only use not found if the database does not provide this information at all.

Note that a database that versions its (meta)data but does not make that versioning public is diminishing the utility of that information to its user community. As such, in those cases, the value of this attribute is no.

Alignment with the RDA

This portion of the FAIRsharing data model was created in response to the needs of our user community, and through a collaboration with the RDA and its relevant outputs. FAIRsharing is also represented on the RDA Data Repository Attributes WG, and will implement the RDA recommendations that will result from this WG.