• View
• Edit
• History
• Print

Preservation Guide - Getting Started

2.1 Where to start: cartography! The very first thing is to know where you are: what’s on the shelves. This information gives you a map of your audiovisual collection.

How to make a map:

The map is not a catalogue. It is not about each individual item, but about each type of item. When a collection has been mapped into types, the preservation planning is then done for just these few types.

Do you know where you are, or is there terra incognita in your collection?

The first step is to decide how to divide the collection into types. The goal is preservation, so divide the collection into categories that have distinct preservation requirements.

The obvious place to start is with physical formats: the different kinds of media. Basically, you can start with dividing the collection according to the machine used to play an item. This would separate film from videotape, and separate U-Matic from Beta SP – but would not separate 5” reels of 6mm audio tape recordings from 10” reels. More detailed information is in PrestoSpace Deliverable D5.3 Tools for automation of difficult media. There is also abundant online technical information about film from Video Aids to Film Preservation, which includes short training films that can be played online, or downloaded (under a Creative Commons licence).

[What this wiki needs is a whole section on technical information, although much could be done by simply pointing to existing online information --- but pointing in an organised way.]

How old is each format? If you have materials from a range of ages (such as 1980 to 1985) then it would be very useful to know how many items from each year. If you have separate age ranges (such as audiotape from the 1950’s, and also from the 1980’s), then consider breaking up the physical format into two age types (or more if needed).

How has the material been stored? You may not know, or there may be various storage histories. Possibly some have been stored in an archive (with climate control) for most of their life, and others have been stored in home or office conditions for most of their life. Again, it would make sense to divide such a physical format into two storage types: controlled storage vs non-controlled storage.

Genre: So far, we haven’t asked anything about the content! We’re not interested in preserving blank material, so content matters. Under the term genre we create a set of types according to content. Typically types are: fiction or non-fiction; news, entertainment, documentary, drama. Again, the goal is to simplify and organise the decisions about preservation – and there is no need to preserve material of no value. So the label ‘genre’ is really for anything about the value of the content.

Once the types have been decided, the next step is to evaluate the collection – get information on each type.

1. Divide the whole collection into physical format types
2. Count the number of items in each format type (counting shelves is the usual method)
3. While counting shelves (step 2), estimate the age of each type – not the age of each item. Where necessary, you will already have subdivided the format type into age groups (ie 1950’s audiotape and 1980’s audiotape), so while counting shelves, estimate the number of items in each age group – for each format.
4. Another task while counting shelves is to estimate the storage history (life history). Again, where necessary, you will already have subdivided the format type by storage conditions as needed (eg items that have mainly been in controlled storage vs items that have mainly been in non-controlled storage).

So far, we’ve made a big assumption – that your materials are already arranged on shelves divided according to physical format, age and storage history. If this is roughly true, the mapping becomes a shelf-counting exercise. If your materials are not divided this way – for instance if material of different ages are on the same shelves – then you need other ways to get the needed information. Possibly age information is in your catalogue. Possibly everything you need to know for preservation is in your catalogue (in which case you are unique, because general experience is that preservation work exposes everything that isn’t in the catalogue!).

If your shelves don’t divide nicely by format, age and storage history (simultaneously), it may be necessary to count shelves more than once, separately for each relevant factor (like age or storage history).

Now comes the crucial part: physical condition. Here you will probably have to use sampling, because it would take far to much time and money to test every item. Alternatively you may have information already, from your own experience or from the experience of users of the collection. The basic task is to estimate the degree of playback difficulty, and the life expectancy, for each grouping. If possible, that means for every physical format, and for every division that you made according to age and storage history. Testing by genre shouldn’t be relevant, because playback and life expectancy depend upon the physical factors, not on the content.

• test whether an individual item plays first try; if it does not, record how much effort in hours (cleaning, baking, trying other equipment) is needed to make it play.
• assess the general condition and life expectancy. For acetate-based materials there is a simple method to check the acetic acid level and predict life expectancy. That procedure is fully explained by the Kodak Image Permanence Institute refs?. Other aspects of condition checking are covered on various websites refs …?.

The result of this exercise is a new description of your collection. It is largely a physical picture (except for the genre / value information), it is based on a small number of categories rather than on individual items, and it supports decisions about how to proceed with preservation work – arranging the categories in a priority order (like the triage in an emergency ward) and then proceeding with preservation category by category – a divide and conquer approach.

Sample map:

As you can see, this film collection has been divided into five parts for the purposes of preservation decisions. This data is based on a subset of the BBC’s film holdings – and does not cover 35mm film or even16mm colour film apart from Ektachrome, and does not cover other complexities of film like “A-B rolls” and “internegs”. So it is really a corner of the BBC film map, which has other maps for audio and video materials. As you can see, the format “16m mag sound track” was divided into two types: masters and dupes. This was recorded under genre, because the essential difference was one of value. A duplicate is of little value providing there is a good master.

What the map shows is that the 16mm B&W collection had to be divided into four parts: negatives, prints, sound track masters and sound track duplicates. For investment in preservation, we use the value table to decide to do nothing about prints and sound track copies – and concentrate on the other two categories. We then had to take immediate action on the vinegar syndrome materials, but have deferred further action on the B&W negatives. A detailed map would include life-expectancy estimates for the B&W negatives, showing how much time is available before further preservation actions are mandatory.

Further information: A map is a simple idea, and a basic map is straightforward. When the categories (the countries on the map) are not clear, ‘border disputes’ can arise or categories may have to be subdivided. Since physical format and physical condition are the important features of a preservation map, technical knowledge is required. This technical knowledge isn’t specifically “preservation knowledge”, but preservation draws on such knowledge. Knowledge about formats, their players, their histories, and about availability, operation and repair of equipment all are relevant. Format obsolescence is one reason for doing preservation work, and so there has to be knowledge about which formats are obsolete – and which will soon be obsolete.

A Supplement? gives general sources of information, including professional organisations that provide training and conferences.

A useful article on ‘preservation cartography’ is: Dominique Saintville, INA (France) - Preparation of legacy archives for digitisation: the INA migration plan. September 2001 FIAT Conference ((locked FIAT reference ; public reference?))

top 2.2 Muster your resources

The map of the collection divides the preservation territory into manageable pieces. Before planning a strategy for the collection and for the preservation work (one or more preservation projects), there is more basic information to gather. Now that there is a map for the collections, information is needed on what resources are available. The map will have identified preservation needs – so now the issue is preservation resources. This won’t be shelf counting, but it could be head counting – and looking at finances.

The following are the basic categories to consider when gathering information on the ability of your archive to undertake preservation work. But – resources to do what? The collection map has only identified broad areas, which is not a preservation strategy or plan. One of the biggest choices in planning preservation is the decision about what work to do in-house, and what needs to be done under contract with a service provider.

There is a chicken and egg problem here. You don’t need a lot of internal resources if a contractor does the work – but in order to make that decision you do need to know what could be attempted in-house.

Do you have what it takes?

• staff: technical staff to operate equipment. Cataloguing staff. Logistics staff to find and move media. Management. If you are a one-person archive, then your time has to stretch to all these activities.
• equipment:
  • in-house preservation work needs equipment: for cleaning old formats, for playing old formats, for writing / printing new formats. A key issue in videotape transfers is headlife: whether there are enough heads (or a reliable source of reconditioned heads) to play however many hundreds or thousands of hours of material to be transferred. For many archives around the world, there is currently (2006) a worrying shortage of equipment for playback of audio DAT recordings.
  • contracted preservation work needs less equipment, but there still may be a need for players for checking as part of quality control.
• facilities: a place to do the work
• logistics: people and transport for finding and moving media as required, and (for transfers) for taking old material out of the collection and introducing new material
• metadata: maybe this should be at the top of the list. An archive runs on its catalogue. The mapping has already shown the importance of data about the collection. If the catalogue or database shows age, storage history, information about storage history -- and user or operator feedback about physical condition and quality – then mapping can be done just from the database itself, without walking around the shelves counting everything. Similarly, if material is bar-coded, and already identified to minimum cataloguing standards – the preservation work will progress efficiently. If not, then time, effort and material will be lost. Finally, if the metadata is missing or not up to standards ((ref FIAT minimum data list; Dublin core )) then completion of metadata should be an essential part of preservation work – which will add to the time and cost estimates.
• existing funding: thus guide would prefer to ask the question: “Can you afford NOT to preserve your collection?” However money is a key issue – and a surprisingly important issue is the accounting procedure. If your collection already has space and staff and equipment that are not fully occupied, a small amount of preservation work can be ‘kept off the books’ because the true costs are invisible (to the accounting process).

This form of ‘preservation for free’ is attractive, but it can be a fatal attraction. If the work required exceeds the resources available, material will be lost. Further, it is hard to get financial approval for doing the required work at a certain price – if it is possible to do a fraction of that work effectively for free. The ‘invisible accounting’ limits the ability to properly plan and fund a comprehensive preservation project.

top 2.3 Prioritise

Preservation has been defined in terms of permanent access, but that definition is the headline, not the detail. There is still more information to collect before starting planning, and that is information about preservation priorities for your collection. The following is a suggested list.

• (keeping the passengers happy) business priorities: what do you do? Preservation decisions should be about purposes, not about media. Fundamentally, the object is preserving access, and so it is highly relevant to ask what kinds of access already take place. Who uses the collection, what parts do they use – and what would you the collection manager want to see for future access?
• (getting a new engine) format obsolescence: this is the area of technology priorities. Looking at the collection map, it should be pretty clear which areas have problems of format obsolescence.
• (keeping the hull watertight) material degradation: this is the area of physical priorities, the condition assessment part of the map. For both format obsolescence and material degradation, the priorities for preservation work change over time – so planning has to foresee the point at which obsolescence or degradation become very high priority.
• (getting new passengers and going to new ports) new business opportunities: what do you want to do? What could you be doing? This area could be essential, because it could be the prospect of new services that opens the door to new funding. Certainly it’s a lot easier to raise funds when describing access and new forms of access – than when describing old material and forms of damage to old material.

An audiovisual collection isn’t exactly a cruise ship, but audiovisual material is in danger of sinking. The cruise ship analogy does show how priorities change with circumstances. Getting a new hull isn’t at all a priority until the old one is in danger of leaking – and if it does start leaking then keeping the water out doesn’t just get a higher priority – it becomes the sole priority.

top