CERN-ETT-2001-002 by Anita Hollier - April 2001
CERN Operational Circular N° 3 "Rules applicable to archival material and archiving at CERN" describes the functions of the DRO as follows:
"Each division leader nominates in an official function a DRO, to whom the authority for implementing a records management and archiving plan within the division is delegated. The DRO is empowered to advise his division on the creation of intermediate divisional archives in order to facilitate a degree of autonomy with respect to records management and archiving within the division. The DRO identifies archival material and informs the CERN Archives prior to restructuring, cessation of activities in the division or departure of staff who possess files of potential interest to the Organization and the CERN Archives. The DRO is periodically invited to report to the Archive Committee."
This manual is intended to give some more practical advice to the DRO about how to fulfill these responsibilities. I hope it is useful. I welcome your comments, questions and feedback - please contact Anita.Hollier@cern.ch.
A. Hollier (ETT/SI), April 2001
- What do you mean by a record ?
- So electronic records should be treated in the same way as paper records ?
- What is the lifecycle of a record ?
- Do we have to keep everything forever ?
- How do I know which records should be sent to the Archive ?
- What items should not be sent to the Archive ?
- What's the procedure for sending files to the Archive ?
- Where will I find the records that should be sent to the Archive ?
The International Council on Archives defines a record as " ...a specific piece of recorded information generated, collected or received in the initiation, conduct or completion of an activity, and that comprises sufficient content, context and structure to provide proof or evidence of that activity."(i)
The information may be recorded in any physical form or medium, e.g. a memo written on paper, a photograph, an electronic document, etc, etc. What they have in common is that they provide evidence of a transaction or activity.
The value of a piece of information is the same regardless of how that information is recorded. It is the content not the medium that is important. However, electronic media present some special challenges because computers make it very easy to create and store information and documents that do not necessarily have the characteristics of a genuine record. Electronic records are also very vulnerable to accidental corruption or loss; and even if they are undamaged, their authenticity can be difficult to prove, so they lack evidential value.
A working group has been set up to address the management of electronic records, and an operational circular will be drafted. The principle is the same as for management of any records, but responsibility is shared between more people. IT specialists will ensure that systems exist which are capable of storing information in the form of evidentially valid records. But it is up to the DRO and the Division Leader to ensure that important records are identified and correctly managed in these systems.
Current records are the documents that are needed for current business. They are consulted frequently, and are best kept in or near the office so they can be accessed quickly.
Semi-current or intermediate records are less frequently consulted, perhaps because they relate to a project or activity that is finished. These can be moved out of the office in order to free space. It is a good idea to set up an interim storage area for your division or group where these sorts of records can be safely stored.
When documents are no longer needed for business or legal reasons they should either be passed to the Archive or destroyed. The choice depends on a number of factors including the importance of the subject matter, and the provenance of the record (see section 5).
No. Not all records have the same value. Some should be kept forever, but some can be thrown away after a few days, and most lie somewhere between these two extremes. It would be prohibitively expensive to attempt to preserve and maintain access to all CERN records indefinitely. It would also be impractical to manage. Even in the electronic environment, where data storage costs continue to fall, the full costs of cataloguing, maintaining, migrating and providing access make it impractical to keep all records forever. It is not in the interest of CERN or its member states to retain records for longer than they are reasonably required to support identified needs.
These appraisal judgements can be difficult, and it is worth investing some time to draw up a retention schedule (see section 10) to ensure such decisions are made in a controlled and efficient way. It will enable you to:
• Minimise the risks of unplanned and unauthorised destruction of records
• Reduce the hidden costs (of both time and money) of continuing to keep records unnecessarily
• Identify more easily and systematically those records that should be transferred to the CERN Archive.
Selection of material to be sent to the Archive is most efficiently done as part of a wider records management plan (see sections 9 & 10). But the following guidelines, together with the definition of archival material in Operational Circular 3 (ii), should be helpful whether or not such a plan is yet in place.
The selection of records for the CERN Archive depends on their historical value. The Archive does not pretend to keep all documents of potential legal value or permanent operational value - though archival records may have both. In general the sort of documents to be kept for historical purposes are those that deal with the making of strategy and policy and those (not including published material) which record scientific and other results. CERN archival records may be divided into three types:
5.1. Main series of key records, produced at Division level or Group level
Reports, technical notes and specifications, minutes of meetings. A list of these can fairly quickly be drawn up. Copies of all documents on the list should be set aside for the Archive as soon as they are produced, and these series should be transferred regularly.
5.2. Other key documents, produced on an ad-hoc basis
The records of ad-hoc Advisory Committees, or the equivalent, are important, as are documents relating to the setting up of experiments, proposals, letters of agreement, memoranda of understanding, contracts, designs and plans.
Budgets and summary financial records, audit reports, post-project appraisals and any other reports or workplans which give a good overview of activities within the area should be kept. Research files, notebooks and logbooks (particularly the "experiment" or "running" logbook kept by the full collaboration) are also worth keeping for the more important experiments. Public education and public relations material, including lecture notes, can be interesting; and organigrams give useful information about the structure of the organisation.
As far as possible, lists of these records should be drawn up in advance. They should be offered to the Archive, via the DRO, as soon as they are no longer required for current business.
5.3. Filing systems of individuals
These collections may contain ANY type of document, including working notes, correspondence, memoranda, drawings, photos, sound recordings, etc. The most important consideration is that they should show the policy, functions, activities and achievements of the body (division, experiment, project, etc) in which they were created. They are important because they give background detail on how decisions or conclusions were reached. Many of the important decisions will be recorded in the key documents listed above, but these can be rather "dry", and don't necessarily give an understanding of the real issues involved. The most historically valuable records will normally come from the higher levels of the organization. For example, the files of a Division Leader or Collaboration spokesperson are nearly always valuable as they include important correspondence and memoranda. The same is true for the leaders of important projects or groups, and other key individuals.
These records should be offered to the Archive, via the DRO, as soon as they are no longer required for current business. If possible, the DRO should separate the more important files from files of day-to-day matters and keep only the former. However, there is no need to "tidy-up" the content of the files before sending them; in fact, it is important that the original order (or disorder) is respected. If good records management procedures have been followed this problem will not arise, but the reality is that many files are a muddle of different documents. In these cases, if a file is mainly of historical interest it should be sent to the Archive in its entirety, even if it also contains some "rubbish".
Different parts of CERN may have other important files unique to themselves. These should be reviewed by the DRO, in discussion, where possible, with the records' creators, to see if they merit permanent preservation. The following questions may provide some guidance in selecting material.
• Do they provide evidence of the origins, structure, functions, policies, decisions or significant operations of the person or body who created them?
• Do they give a good overview of the progress (achievements or problems) of the Division / Collaboration?
• Do they relate to significant changes in CERN's activities, processes or policy?
• Do they relate to a project that led to a "first of its kind" process, activity or result?
The sort of items not required by the Archive include low-level working papers; excessive detail on relatively unimportant subjects; multiple copies or drafts (unless they show significant changes); day to day administrative matters; raw data, or data summary tapes. These are subjective decisions, and it is likely that more records will be kept from higher levels in the Organization than from lower ones. Similarly, more detailed information will be kept on subjects of greater importance to the Organization, and for periods of crisis (good or bad).
Some documents are widely distributed in CERN, and copies are held in many filing systems. The following series are already sent to the Archive by their "Office of Record", i.e. the office that produced them, so further reference copies need not be sent (unless they are extensively annotated):
Council, Committee of Council, Finance Committee, Scientific Policy Committee and TREF documents; CERN Courier and Weekly Bulletin; CERN Research Board Minutes; CERN Annual Reports; CERN Yellow Reports; Experiments at CERN.
Published material and documents intended for publication are not generally kept in the Archive. (They may well be of interest to the Library; if in doubt, contact email@example.com). However, some Divisions do not differentiate between unpublished divisional reports and preprints, so it is the policy of the CERN Archive to keep all divisional reports, even if many of them are also kept in the Library collection.
Collections of key documents (see section 5), such as series of Divisional Reports, should be sorted chronologically, separated into folders by year, and placed in archive boxes.
Where possible, the documents in individual's filing systems should be removed from their files, refiled onto plastic fasteners and placed in archive boxes. Metal fasteners (apart from staples) and plastic wallets should be removed, as they deteriorate quickly and damage the paper.
The boxes should be labelled before they are sent to the archive and a transfer sheet listing all the items sent should accompany them. If you have a classification plan please send a copy of this too. Any background information you can give about the subjects dealt with in the records, or about the person who created them, will be very valuable in ensuring that they are catalogued as accurately as possible. Plastic fasteners, folders and boxes are available from the archivist on request.
Note on electronic records
If a record is stored on the CERN Document Server (CDS) or on CEDAR EDMS, this can be treated as the Archive copy. It is sufficient to send a list to the Archive, describing what records are available where. However, it is not yet certain that documents stored electronically elsewhere (e.g. locally or on the WWW) will be secure in the long-term, so paper copies must be sent to the Archive.
Not all Divisions have a well-maintained central filing system, so until such time as a records management plan has been put in place the DRO may have to do some detective work to find the records of archival value.
The main resources are the Division Leader or Collaboration Spokesperson, whose files should give a high-level overview; and the Divisional or Collaboration Secretary who should be able to supply most of the major series of regularly produced key records. However, the original files and notebooks that deal with the most interesting activities may not be easy to find. The DRO will need to approach individuals to ask for this material. In some cases the DRO will have a good idea who to ask and where to look for historical information. If not, it is helpful to consider what are the ongoing functions of the Division and where in the organisation these were carried out. Even though group names change, and groups are moved between divisions, the same functions usually continue to be carried out somewhere. A memo circulated to Group Leaders explaining what records are wanted and why will help to raise awareness of the issue.
"Archiving policy at CERN", the subsidiary document to CERN Operational Circular N° 3, says:
"Archiving not only involves the preservation of documents of long-term interest, but also requires organized records management and the application of defined standards at the moment of creation of a document."
Records management allows the creation, maintenance and disposition of information to be carried out in a more efficient manner. This saves time and money, for example by reducing the amount of space wasted in storing unimportant records, and the time taken finding important records which are lost among them. It also helps to ensure that the records which must be kept, whether for business, legal or administrative reasons, are kept safely. This is particularly important in the case of electronic records (see section 2). If valid records are not kept, the organization faces a number of risks including:
• Failure to produce evidence that it has fulfilled its obligations
• Loss of proof of ownership or rights
• Inability to find information required for current activities
• Loss of efficiency if this information is difficult to find, and potential high costs of "rescuingc electronic records if they are not in a condition to be accessed directly
• Inability to rely on information.
Good records management will also facilitate the sharing of information, while safeguarding any necessary access restrictions.
You will need some extra manpower to carry out the following tasks. The archivist is always available to advise, but an assistant will also be required in most cases to help with the practical work. It may be helpful, when requesting resources, to present your Division leader with a finite project with clear deliverables. Remind him of his and your responsibilities under Operational Circular 3. Point out the weaknesses of the current system and the advantages to be gained from implementation of a records management plan.
10.1 Analysis of business activity
First, carry out a function-based analysis of your Division or experiment to identify the business processes and information flow, and identify the points for which a valid record is required. In discussion with the relevant Division and Group leaders, you will need to identify:
• The Division's goals and the strategies to achieve these goals.
• The broad functions which the Division undertakes to support and fulfil its goals and strategies.
• The activities or major tasks which contribute to the fulfillment of the Division's functions. There may be several activities associated with each function.
• The groups of transactions or processes which make up each of these activities. Transactions are "the smallest unit of business activity". They should be activity-based rather than subject or topic-based. Transactions provide the basis for identifying, in detail, the records that meet the business needs of the organisation.
It is important to identify all points at which records are (or should be) created. These should be designated as Official Records, which must be properly stored.
Deliverable: a process / data flow model showing processes and data stores.
10.2 Records inventory
Next, with the aid of an assistant, survey the information currently held and draw up an inventory. This should identify significant groups of records, and for each group describe:
• The title of the record series
• The dates
• The purpose for which the records are kept
• The current arrangement and reference system
• The media on which the information is stored
• The quantity and rate of growth
• Whether or not they are the "record copies" (see section 11)
• The storage location.
If information is available about the legal requirements for retaining the records this should also be noted.
The most reliable way to produce the inventory is by a combination of interviews with the relevant staff (based on questionnaires) and a physical survey. The inventory also serves to check that the process model created in step 10.1 is accurate. All the records series and data stores found should fit into the model.
Deliverable: a survey of the records series that are currently kept, both in paper and electronic form.
10.3 Retention Schedule
Compare the inventory with the process model and use them as a basis for a discussion on information requirements with the information creators and users, the Division Leader or collaboration spokesperson, and a representative from the Legal Service. Next, draw up a retention schedule that specifies the length of time for which each series of records must be kept to meet legal, operational or historical requirements. If the same record is available in different media, it should specify which should be kept. Specify the confidentiality and access requirements for each group of records.
Deliverable: authorised retention schedule, signed by the Division Leader, after consultation with the Archivist and one of CERN's legal advisors.
10.4 Classification plan
If necessary, draw up a classification plan for all records, identifying a hierarchy of classes of records based on the business procedures, and selecting the most suitable coding scheme to create a unique identifier for all records (see section 11).
For the numbering of CERN scientific documents, see Operational Circular 6 (in prep.)
Ensure that the relevant staff are familiar with the retention schedule and that records marked for permanent preservation are transferred to the Archive at the correct time.
Non-permanent records should be destroyed when specified by the retention schedule, bearing in mind any confidentiality considerations. The inventory and retention schedule should be regularly checked and updated
Setting up a filing system (or revising the one you have)
Because creating, using and storing files takes a lot of your valuable time and uses expensive office space, it should be your goal to save each record only as long as it is really needed
An efficient filing system serves both short-term and long-term needs. It increases the accuracy and efficiency of handling information throughout the lifetime of the files, and makes it much easier to identify the records worthy of permanent preservation.
Know the responsibilities of your office
It is worth spending a little time identifying the responsibilities of your office - what jobs you do, and for whom.
Identify the documents that reflect the work of your office
Once you have identified these responsibilities, make a list of the kinds of documents (letters, minutes, reports, receipts, etc, relating to different functions) produced in the course of each of these activities. Also list the kinds of documents produced elsewhere that you and others in your group currently need to do your work (this is "non-record" material, see next section). Strictly personal papers should always be filed separately.
Distinguish between record material and information copies
Among the documents produced by your office there will be some that no other office has, such as letters, informal notes on meetings, unpublished internal reports, etc. These are known as "Record Material" and your office is the "Office of Record" for these documents. Even if copies are sent out to other places, your copy is the record copy. For example, if you keep files for someone who chairs a committee, your copies of those minutes and agenda are the record copies, and all others are non-record or information copies. Non-record material should be weeded (sorted) regularly, at least once a year, to ensure that only current and useful items are kept.
Create a series of "record material" files
It is suggested that you create a series of files that contain just one virgin copy of all the important record material related to the work of your office, i.e. the key documents. Duplicates for distribution should be stored elsewhere. Working copies may also be filed elsewhere with any relevant non-record material that is needed for current business.
In general, non-record material need only be kept for as long as it is needed for reference, and keeping record material separate from the beginning eliminates time-consuming weeding when it is time for the records to be transferred from your office.
Break down the files into series
Create logical groups of files that relate to a particular subject or result from the same activity. Ensure that all the files have clear titles and an opening date. Add the closing date when the file is full.
Create a classification system
All but the smallest filing systems will be easier to manage if a standard classification system is used. There are many possibilities, e.g. alphabetical, numerical, chronological; you should choose the one that suits your needs best. There is no point in developing a complicated numbering system if you keep case files on individuals and generally need to retrieve records by a person's name. In this case an alphabetical system would be easier, perhaps supported by colour coding when the collection gets too large. Another common way of classifying files is by Function, Activity and Transaction. If you would like further advice, contact the archivist or your DRO. Mark the classification reference clearly on each file, and keep a copy of the classification scheme nearby for easy reference.
For the numbering of CERN scientific documents, see Operational Circular 6 (in prep.)
• File regularly to avoid confusion.
• Keep papers in chronological order within files or sections of files - with the most recent items on top.
• Don't include multiple copies.
• Avoid using paperclips, adhesive tape or plastic folders in files that are likely to be sent to the Archive. These materials deteriorate quickly and damage the papers.
(i) Conseil international des archives - International Council on Archives. Guide For Managing Electronic Records From an Archival Perspective. Committee on Electronic Records, February 1997. (ICA Studies/Études CIA 8) Section 2.1
(ii) From CERN Operational Circular N° 3, Rules applicable to archival material and archiving at CERN
Definitions and general remarks
A - Archival material [Top]
- Archival material is any documents originated or received by a person or a body in the course of activity in an official function at CERN, which is preserved in the Archives because of its long-term interest, regardless of its physical form or technical support.
- Files or documents setting out the policy, planning and operational structures set up to carry out and support scientific research are of great importance both to the Organization and for scholarly research and must therefore be archived in a manner which facilitates access, this avoiding costly retrieval procedures.
- Files or documents which fall into the following categories have to be preserved:
- CERN Council and its subordinate bodies.
- Administration, management and policy-making.
- Accelerators: planning, construction, operation.
- Research facilities, experiment collaborations.
- Theoretical physics.
- Technologies: computing, electronics, vacuum, cryogenics, etc.
- Spin-offs: technical, economic and cultural.
- Training, conferences, utility studies , etc.
- Cultural and social life at CERN.
- Unpublished CERN divisional reports, technical notes and specifications are archival material.
- In keeping with the provisions of CERN Administrative Circular N° 29 on CERN publications, documents intended for publication are, as a rule, not archival material.
B - Special archival collections
- Special archival collections at CERN include:
- Administrative files to which special obligations apply (Personnel, Finance, Medical Service, Pension Fund, etc.), as defined in Administrative Circular N°10.
- Audio and video recordings.
- Photograph and films.
- Press articles.
- Technical drawings and plans.
This type archival material is, at present, the responsability of the units concerned and is subject to special agreements between the originators or the persons responsible for these units and the CERN Archivist.