Archives and your Personal Papers

Martin Campbell-Kelly & Simon Lavington
Over their working lives many computer professionals acquire materials of potential historical value. How do you ensure they are preserved? There are no hard and fast rules, but here are some ideas to get you started.

First bear in mind that archiving is costly. An archive takes in materials with the aim of storing them indefinitely. It provides access to scholars and researchers and deals with copyright issues (for this reason donors are asked to assign copyright to the archive, and also disposal rights so that unwanted materials can be removed). Maintaining an archive is expensive both in terms of on-going cost, and in “opportunity cost” because archives are finite, and saving one set of papers means that another will eventually have to be foregone. Unless your papers are somewhat unique they are unlikely to be of interest to an archive in total, but some items may well be of value. If you really want your papers preserved, consider providing some funding to the archive: this won’t affect the decision to accept materials, but it will speed up their processing.

Second, you need to get your papers in order. Unless your papers are self-evidently important no archive is likely to accept an unsorted box of miscellaneous materials. Getting your materials in order involves removing non-archival materials and putting the remainder into folders in some kind of systematic order (often chronological). It is useful to add a note to explain the contents of a folder at the beginning. If you can produce a hand-list or catalogue as a free-standing document, it will enable an archive to appraise and process the collection much more easily. Also, it may help subsequent researchers if you annotate (with initials and date) obscure items. It is unlikely that your heirs will be able to do this for you.

Third, archival storage in the UK is in short supply compared with, say, the United States. The principal UK repositories for computer history are the National Museum of Computing, the National Archive for the History of Computing at Manchester University, and the Science Museum. There are also less obvious places: for example, museums and libraries often maintain local history archives – if you worked for a local company there may be some interest; universities commonly maintain archives and may be interested in small collections. There are also specialist collections for significant companies: for example the Marconi Archive (which includes Elliott Brothers) at the Bodleian Library, Oxford, and the Ferranti Archive at the Museum of Science and Industry, Manchester.

What Materials will Archives Accept?

Generally, computer archives are most interested in items that are 25 or more years old. Since the PC era, there has been an explosion of materials that archives have inadequate resources to deal with. Hence the interest is mainly in items that have survived for a few decades (itself an indication of potential historical value) and that are in danger of being lost. No society can keep all the printed matter created, and archives rely on time to winnow out the dross.

Archives are especially interested in “grey materials”. That is, unpublished materials such as notebooks and working papers, research reports, program documentation, sales manuals, etc. Also of great interest are correspondence and memoranda, and ephemera such as product literature and company annual reports.

Computer manuals are problematic. Manuals for first and second generation machines are appreciated if the archive does not already have a copy. Manuals for third and later generation machines are difficult to deal with because of their volume. For example, it was once said the IBM was the largest publisher apart from the US Government: several thousand manuals were published for the IBM System/360. No archive can afford to house so much material which will be of interest to very, very few individuals. For obvious reasons, archives are much keener to accept this kind of material in microfiche form. Bitsavers.org is dedicated to preserving computer related documentation.

National archives are primarily interested in domestic materials. For this reason, archives in Britain will be much more interested in items relating to British computer developments. It is generally safe to assume that US archives will look after their national interests, although all archives will be interested in non-national materials that are both significant and in danger of destruction.

Archives are primarily interested in unique materials. If a book has an ISBN or a copyright notice from an established publisher, then it will be in a copyright library and an archive does not want to store duplicates. For any books published after 1960 the best thing is to give them to a book charity. Some pre-1960 computer books are quite rare, however, and these are worth keeping. Virtually all major periodicals now have on-line access, usually of their entire back catalogue, so are of little value to an archive. Magazines such as Byte and PC World are readily available in libraries so they too are of little interest. However, complete runs of a magazine may be welcome.

Important exceptions to the general rule about published materials are trade magazines such as Datamation and Dataweek, product magazines such as Lotus Magazine, or subscription-only publications such as Software Magazine. Very few libraries have holdings of these, and they could be a valuable addition to an archive, especially if there is a substantial run. Another genre of great importance is reports by industry analysts such as Gartner and INPUT – typically market surveys or products analyses. Very few of these have survived and they are particularly useful historical sources.

Photographs are much sought by archives: they occupy little space and have great historical value. To be of use, a photograph must be accompanied by sufficient information to identify the context and people – if this information is written on the back of the photograph, it cannot go astray. Digital photographs need to have the same information included as meta-data. Movie films and slides are historically valuable, and have few preservation problems.

Software preservation is a minefield and current advice is necessarily contingent. Generally, few archives can accept code per se, but may well be interested in documentation. There are very few examples of programs for which a complete history exists – from specification to deployment – and this would likely interest an archive. Mass-market software may be of interest, if preserved in the original packaging (but note the “25 year rule” above).

Digital media is another minefield. Archives are not yet equipped to deal with digital artefacts except on an ad hoc basis. Generally they cannot hold magnetic tape, hard disk drives, floppy disks, tape recordings, VHS or DVD because all magnetic storage systems have a finite life and data has to be periodically transferred to the latest secure format (currently NAS or cloud storage). If you have an important email archive, a reasonable strategy for its preservation is to print it out selectively on paper. Archives are conservative institutions and they are taking time to adapt to the digital world.