Editorial archives

Journals provide long-term archiving for researchers’ published articles. However, these are not the only documents produced during the scientific publication process. Each publication is peer-reviewed, which involves exchanges between the reviewers and the authors. What happens to these documents after publication? How long should a magazine keep them on file to protect itself against any recourse? How to store them? Can they be deleted? What should be done with the personal data contained in these documents?

These documents are referred to as “research archives,” which are defined as such by the National Archives:

“Records relating to research carried out with public funding are classified as public records. Those held by national research institutions are transferred to the National Archives for historical purposes.

These archives comprise all documents and data created or received during research activities, both in relation to the research itself and its administration. They include the work produced by researchers:

• experimental data;
• photographs ;
• field notes;
• manuscripts ;
• publication of results.

These archives also include records from laboratories and research centres (activity reports, funding applications, correspondence, research protocols, etc.) »

Note:

In the absence of a specific legal text, we will try here to provide practical answers to these questions by looking at both the RGPD and archive management issues. Mais avant tout développement, il convient de préciser la notion de cycle de vie d’une archive.

Archive life cycle

Personal data

Exchanges between the publisher, the reviewers and the authors contain personal data: surnames, first names and e-mail addresses in most cases. As such, these documents are covered by the Règlement Général sur la Protection des Données (RGPD), a French law on data protection.

To find out more, see our article on personal data processing policy.

Life cycle

According to the CNIL and the Direction générale des Patrimoine, in their Guide pratique : les durées de conservation:

The life cycle of a dataset can be divided into three successive phases:

• Everyday use (‘active database’): this stage concerns the everyday use of personal data by the departments responsible for processing it. In practical terms, this corresponds to the files used on a daily basis. The data is accessible, within the immediate working environment, to all those responsible for day-to-day operations.
• Intermediate archiving: personal data is no longer used to fulfil the original purpose (‘closed files’) but is still of administrative interest to the organisation (for example: managing potential legal disputes, etc.) or must be retained to meet a legal obligation (for example: invoicing data must be retained for ten years in accordance with the Code of Commerce, even if the data subject is no longer a customer). The data can then be accessed on an ad hoc basis by specifically authorised persons.
• Final archiving: this refers to data that is archived indefinitely. This phase only concerns processing carried out for archival purposes in the public interest. This last phase essentially concerns the public sector subject to the requirements of Book II of the French Intellectual Property Code.

Thus, in our case, the current use of personal data contained in evaluation reports corresponds to the article evaluation phase, when the journal needs to exchange information with the reviewers.

Intermediate archiving corresponds to the period when the journal is likely to face litigation. In this case, the journal team will need access to the evaluation reports and the different versions of the article.

Final archiving is possible when there is no longer any risk of claims – or when this risk can be considered negligible.

To summarise:

Types of archives	Access	Life cycle
Everyday use: documents used on a daily basis	Documents and data must be easily accessible to the editorial teams managing the process	“Article review process • Submission –> Publication”
Intermediate archiving: Data that is no longer used on a day-to-day basis but is still of administrative interest or must be retained to meet a legal obligation	Documents may be accessed on an exceptional basis, subject to justification, by specifically authorised persons	“Period during which the journal may face legal action • Publication + X years?”
Permanent archiving: data archived indefinitely	Once the data has been sorted, access to it will be restricted to a very small number of people	When there is no longer any risk of legal action, or when such a risk can be regarded as negligible

What the law says

Heritage Code: General Rules on Archives

Articles L211-1 to L214-10 set out the general rules governing archives and the regulations applicable to them. Although the specific case of scientific archives is not addressed in detail, they nevertheless fall within this general framework.

See the general rules governing archives: Art. L211-1 à L214-10

Guidelines for managing research archives

Theréférentiel de gestion des archives de la recherche(guidelines for managing research archives), drawn up by the Association des Archivistes Français and recommended by the CNRS for document conservation, are based on the following texts:

• Book II of the Code du patrimoine.
• Décret N° 79-1037 du 3 décembre 1979 modifié concerning the competence of public archive services and cooperation between administrations for the collection, preservation and communication of public archives.
• Instruction DPACI/RES/2005/003 of 22 February 2005 on the sorting and conservation of archives received and produced by departments and establishments involved in national education.
• Instruction DAF/DPACI/RES/2007/002 of 15 January 2007 on the processing and conservation of the archives of CNRS delegations and the archives of research and service units.
• Prime Minister’s circular of 2 November 2001 on the management of archives in the State’s public services and establishments (NOR : PRMX 0105139C)

It is intended as a guide and provides advice on how to handle and preserve ‘the archives of research laboratories, researchers and teacher-researchers in universities, research bodies and health agencies’.

For example, prooevaluation reports are included in the ‘Scientific production: dissemination of results > Communication of results and preparation of publications (books, articles, etc.)’ section:

• T: Sorting, i.e. destruction or deposit in archives.
• C: Final storage in the archives, with a receipt of deposit drawn up.

A period of 10 years has been set for documents that may be useful in the event of legal action, based on the French statute of limitations:

« – The limitation period for ‘public prosecution in respect of criminal offences’ is set at 10 years, except in the case of offences committed against minors, where the period is 20 years, commencing from the time the victim reaches the age of majority (Article 7 of the Code of Criminal Procedure)

– Civil limitation period: ‘An action for damages arising from an event that has caused personal injury, brought by the direct or indirect victim of the resulting harm, is subject to a limitation period of ten years from the date on which the initial or aggravated injury stabilises’ (Article 2226 of the Civil Code)»

RGPD

What’s more, evaluation reports and the various author documents contain personal data (first and last names of authors and reviewers, emails, affiliations, etc.) and must therefore comply with the General Data Protection Directive (RGPD). The journal must therefore justify the storage of this data and define a legal storage period.

Possible reasons for archiving:

According to Chapter II, Article 6 of the General Data Protection Regulation (GDPR), the processing of personal data (archiving constitutes data processing) may be justified on the basis of six legal grounds:

• Processing is necessary for the performance of a contract
• Processing is necessary to comply with a legal obligation
• The processing is necessary to safeguard a person’s vital interests
• The processing is necessary for the performance of a public service task
• The individual has consented to the processing for one or more specific purposes
• Processing is necessary for the purposes of the legitimate interests pursued, unless the data subject’s fundamental rights and freedoms or interests take precedence.

Please note: the last point does not apply to processing carried out by public authorities in the exercise of their official functions.

In the context of scientific research, the processing of personal data will be based on one of the following legal grounds:

• The processing is necessary for the performance of a public service task
• The individual has consented to the processing for one or more specific purposes
• Processing is necessary for the purposes of the legitimate interests pursued, unless the data subject’s fundamental rights and freedoms or interests take precedence (applicable only where processing is carried out in the context of research, but outside the scope of a public service remit, for example in the context of a partnership with a private-law organisation).

Important: it is best not to rely on consent as a legal basis. It is necessary to be able to prove that the journal has obtained the person’s consent and that they may withdraw it at any time.

Why keep records?

Why preserve?	For what reason?	Who is it for?
Control risks and anticipate legal issues (storage obligations in the event of a dispute, etc.)	Proof value	For researchers/authors
Issues relating to scientific integrity (maintaining the traceability of changes)	Check value	For researchers/authors
To preserve a record of the editorial process and the publisher’s activities, as well as the research currently being carried out = a testimony to an organisation’s scientific activity at a given point in time	Heritage and archival purposes	For historians

Which documents should be archived?

At the end of the intermediate archiving period, the question of sorting arises: which documents or data should be archived?

Whilst retaining all the data may seem unrealistic, sorting through and consequently deleting certain data may also result in the loss of a significant part of our scientific heritage. This is therefore a matter that needs to be considered at an early stage so that the sorting process can be carried out in a rational manner and in accordance with pre-established rules.

What data should be archived? All data relating to publishing activities:

• the different versions of the article,
• the reports,
• the discussions between the editorial team and the editors, reviwers and all those involved,
• other documents, once selected…

So, we can prioritise archiving:

• All data relating to a submission for management by OJS
• All emails relating to a submission for email management

Defining a storage period

The guidelines for the storage of research records recommend that documents relating to the preparation of a publication be retained for 10 years. This corresponds to the intermediate archiving period. This period is indicative, and the journal is therefore free to decide on a different retention period.

Then there is the question of final archiving: what should be done after the 10 years of intermediate archiving?

Analysis grid

If the journal itself wishes to define the storage period for evaluation reports and correspondence between authors, reviewers and the publisher, it is necessary to examine the context, the law, the use of the document, its interest, etc. in order to determine the most appropriate treatment.

The document from the CNIL and the Direction générale des Patrimoine, the Guide pratique: les durées de conservation, contains an analysis grid for identifying the applicable storage periods. This summarises all the questions to ask when determining the period applicable to a processing operation.

This grid looks at the different stages of archiving and the context in which data processing takes place.

Questions to consider

What durations for what purposes?

• Day-to-day management of the journal (life cycle: day-to-day storage)

• Intermediate archives + how the journal operates –> Should I retain certain data to protect myself against potential legal disputes? Which ones? – Limited access to be defined
• How long can I take legal action?
• Archival purpose (permanent records) Limited access to historians and archivists? To be deleted?
• Do I have any legal obligations to keep data for a certain period of time?

Archiving process

• Who is responsible for archiving?
• Who does the sorting?

• What information needs to be archived? For how long?
• What are the rules for deleting data?
• What are the rules for data archiving?

Access to the archives

• Where should you file it?
• What access levels are available to new members of the editorial team?
• What rights of access do people who leave the editorial team have?
• Do you need to restrict access to these archives? Why?
• What is the procedure when communication takes place via email?

We can also add, in relation to the RGPD :

• Have I obtained non-anonymous consent from reviewers and authors to put review reports online?

In practice, what can be implemented at the centre Mersenne ?

Publication of reports

Some journals have decided to make their evaluation reports public. They are then published on the article page and stored under the same conditions.

See, for example, the journal Open Geomechanics, which publishes its peer-review reports: https://opengeomechanics.centre-mersenne.org/

Archiving and/or destruction

During intermediate archiving, documents must be accessible quickly. You also need access to the personal data they contain: names of reviewers, name of the author, etc.

At the end of the intermediate archiving period, what should be done with these documents?

• if it is considered that the limitation period has been exceeded, the documents can be deleted.
• if you wish to keep them for the history of the journal:

– deposit documents in a database and make them accessible after anonymisation (Data.gouv.fr, Zenodo, Nakala…)

– store the documents on a secure server, without anonymising the data.

To summarise:

Types of archives	When?	Where?	Access
Current archives	From submission to publication	An editorial management tool or messaging service, depending on the journal	Without Limits (editorial team)
Intermediate archives	Publication + X years	Archiving tool for journals using OJS	Restricted access defined by the publisher (editor-in-chief, editorial secretary)
Final archives	Unlimited duration	Sorted by publisher	Restricted access defined by the publisher (historians, archivists)

Points to survey

Cost

Storing data on a server costs money, whether the cost is assumed by the journal or the community. Storing all the information can lead to space constraints on the servers and high storage costs.

Information overload

The accumulation of information can lead to ‘information overload’, that is an excess of information that hides the important details. If searching for information becomes too complicated, there is a risk that the search will be abandoned or, conversely, that an excessive amount of time will be spent on it. Sorting and organising information therefore helps to reduce the time spent searching for it and improves efficiency and relevance.

Digital pollution

Data storage has a significant impact on the environment.

Firstly, servers run continuously. Powering them requires large amounts of energy. According to Ademe, “The digital sector accounts for 4.4 per cent of France’s carbon footprint”. In the same article, Ademe points out that “even when a server is in standby mode, it consumes an average of 100 watts so that it can answer requests instantly, and a large data centre contains thousands of them”.

Furthermore, apart from energy consumption, data centres are major consumers of water. This is because servers generate heat, and maintaining them requires cooling to prevent damage. Large quantities of energy and water can therefore be used to cool the servers. According to Ademe: “In addition to energy consumption, these cooling techniques require, depending on the method used, a large quantity of water – a precious resource that is becoming strategic due to the effects of global warming and the resulting water stress”.

The EcoInfo website offers an introduction to the environmental impacts of digital technology, which sets out in detail all the ecological impacts of digital technology.