Report urges scientists to secure their records

31 October 2001

A new report underlines the importance of keeping scientific records to safeguard our scientific heritage. Once an experiment is complete, its responsibilities are not over.


According to a recently released report by the American Institute of Physics (AIP) Center for the History of Physics, there are many problems facing the documentation of collaborative research. These range from the way in which the contributions of distinguished individuals (or records of a project conducted by one institution) are preserved, to the fact that, almost without exception, research institutions and federal science agencies fail to provide adequate support to programmes to save records of significant research.

To help to find solutions, the AIP History Center has issued Documenting Multi-Institutional Collaborations – the final report of its decade-long study of multi-institutional collaborations in physics and allied fields.

The main recommendations of the report are that:

* scientists and others should take special care to identify past collaborations that have made significant contributions;

* research laboratories and other centres should set up a mechanism to secure records of future significant experiments;

* institutional archives should share information.

The long-term study focused on high-energy physics, space science, geophysics, ground-based astronomy, materials science, medical physics, nuclear physics and an area called computer-mediated collaborations. The main goal of the project was to learn enough about these transient communities to be able to advise on how to document them.

The study was built on interviews with more than 600 scientific collaborators; numerous site visits to archives, records offices and US federal agencies; and advice from working groups of distinguished scientists, archivists, records officers, historians and sociologists. The study group gathered and analysed data on characteristics of collaborations, such as their formation, decision-making structures, communication patterns, activities and funding.

According to the report, scientists in multi-institutional collaborations are well aware that their way of doing research is unlike that of others working alone or in small groups. All too often, however, scientists fail to realize how records needed to document research are prone to destruction. It may appear to them that their recollections and those of their colleagues are sufficient. This is thought to be unfortunate from the standpoint of present needs. From the standpoint of the future it is disastrous, for even the imperfect recollections will die with the scientists and later generations will never know how some of today’s important scientific work was done. For particle physics, the report has some specific suggestions.

Core records

The report makes a broad distinction between “core records” — those records to be saved for all collaborations — and records to be saved for “significant collaborations”. The definitions of the former are slanted towards traditional US procedures with Department of Energy or National Science Foundation funding for experiments carried out at major US laboratories. However, these can be paraphrased unambiguously for a more global audience without too much trouble.

The additional records for significant collaborations include correspondence between the experiment spokesperson, the experiment collaboration and laboratory administrators. Intracollaboration meetings, collaboration groups, interinstitutional committees, and project management and engineering documents are also deemed to be important under this heading.

bright-rec iop pub iop-science physcis connect