Selection Criteria for Digital Collections

            Libraries of all types, of all sizes, in every corner of the globe, are jumping aboard the digitization bandwagon – throwing up anything they can slap on a scanner to give themselves a web-presence.  But if this mad rush to join the “Digital Circus” (as Sally Buchanan calls it) is to be meaningful, if libraries truly wish to add valuable digital projects to their collection development, then careful thought must be given, selection criteria set down, and serious attention paid to those who have written and thought about the issues.  Selection is much more than just picking what items to convert from analog to digital – it is a complex number of decisions informed by the project’s (and organization’s) mission, and dependent on criteria which should take into consideration such aspects as value, use, copyright, conditions, costs, and relationship to other projects.  It is not a simple task with traditional mediums, and is no easier in the digital arena.   

            Even before selection criteria can begin to be thought about, a popular misconception should be cleared up.  Many libraries are undertaking what they are calling “digital preservation” projects.  In fact, what they are actually doing are “reformatting” projects.  The phrase “digital preservation” has been thrown around carelessly.  It is at the best inaccurate, it can be hurtful by setting the library world up with a false sense of security, and at the worst it is fooling us into a non-truth that could eventually spell disaster.  There is, as of now, no real method or practice for the long term preservation of digital information (though a few brave souls are working on it).  A library is not preserving a book by scanning it.  Digitization, for now, must be thought of as a matter of access and value added reformatting, not preservation.  However, this does not mean that we should ignore preservation issues!  Preservation matters should be of concern at all times during the selection process, as well be considered throughout all phases of digitization (as with any collection development).  Although this is not a criteria per se, understanding that digitization does not grant immortality (yet), will help a library when considering collections for digital projects.  Thinking about digital preservation problems such as media decay and technological obsolescence will also help the library in setting policy for insuring that whatever digital forms the new collections become, they will be carefully taken care of (refreshing, migration....) long into the future.  This is also important for libraries to realize because it will help them think about putting items back into the preservation loop after being digitized, instead of discarding the original or just thinking that it’s safe because it’s on disk somewhere.

            Though some writers and thinkers on the issue of selection like to rank their criteria, it is important to see the criteria as part of an integrated whole of the selection process and to see selection in the context of the whole digitization process.   All of this should be aligned with the philosophy of the organization and the mission of the project. With that in mind, the first major issue to consider is value.  Value is a broad, and subjective term that includes artistic value, economic value, and intellectual value.  The questions to be asked are: Does the access provided by digitizing this collection give value to the community?  What is the purpose of this project?  What added value does digitizing this collection bring?  Added value is an important concept, because if a library is going to spend time and money to digitize a collection, it should be to provide the user with a greater value than was available through the original.  For example, this could be in access to the item, or in search and/or manipulation capabilities.

            Connected to value, is the next criteria, which is use.   Who are the intended users?  Local? World-wide? Both?  Will they actually use this collection?  Is there a demand?  What will the anticipated use be?   Knowing your users, and how they will use your collections cannot be overstated.  Here is where a well devised user-survey could be of great use.  What do the users want?  How about a nation wide survey, perhaps organized by ALA, or RLG, administered locally, that would ask users of libraries what they felt would make good digital collections.  An interesting issue with users of digital collections is they are no longer just local -  libraries have always attempted to serve their community, but now because of the communications revolution we call the internet, a library’s community can be the whole world.  This too must be taken into consideration when thinking about use as selection criteria.

            Copyright is one of the stickiest of all issues for digital projects.  People like Dan Hazen go as far as saying that copyright should be the first criteria to be looked at before anything else.  If the item is not in public domain, and the library cannot obtain permission, then that item is not legally allowed to be used.  Others have listed copyright in a separate category from criteria, like “constraints”, or “preconditions”.  Hopefully, sometime in the next few years, there will be some changes in copyright and intellectual property right law - a revised law that allows for libraries to grow with the internet, and take full advantage of what this digital revolution can allow for our knowledge centers (something like Copyleft). But until that fight is won, libraries must be careful to avoid copyright entanglements by taking such matters into consideration before the fact.  Copyright also presents another problem which is that a library may have a collection that resides mostly in public domain give or take a few items.  In this case, the library must be sure to leave easily read documentation for the user to explain gaps in the collection.  The fact that only public domain material can be posted on the web has been a factor in the widely held misconception by young students that if it’s not on the web, it’s not out there.  Libraries must take measures to inform users of what is held in their physical collections as well!

            The physical condition of an item, and/or collection, is an important criteria for several reasons.  First of all, the condition of an item – including its size, shape, colors.... - will determine how it will be digitized.  The condition of some material may make it a better candidate for preservation techniques like microfilming, especially if its life span is short.  Condition is also important to think about after the digitization is done.  What will become of the original?  Will it be returned to circulation?  To the preservation department?  Discarded?  When thinking about condition as a criteria for selection, libraries should also keep in mind criteria related to condition, like the time and cost involved for material that is in extremely bad shape.

            Costs are of course a very important criteria for selection.  It is the fact that money and time are limited that forces libraries to make selection decisions for their collections in the first place.  Costs are related to everything and should be considered in such a way.  This includes financial costs like overhead, staff labor, equipment (hardware and software), and long term maintenance of infrastructure.  However, this also includes

not so obvious financial costs – for example: is the money going to this digital project draining resources (money, staff) from other important library functions?  How does the cost compare to the benefits?

            The final important criteria for selection to consider is that of the relationship of the project to that of other institutions.  For some reason, libraries that were very involved with cooperative efforts with the last great reformatting period of microfilm, are not talking to one another as much today.  This is too bad, and hopefully will change.  The point is that sharing of resources works extremely well in the library world, and so partnerships should be formed.  When a library is deciding upon what to select, the decision should not be made in a vacuum!  This will prevent doing work that has already been done, as well as help forge relationships between institutions to work on new projects jointly.

            As well as those main criteria for selection, libraries should not forget to think about the issues of staffing and technology.  Staff are a libraries most important resource.  They are also the most expensive aspect of most projects.  Will the library have to hire more?  Do they have the right skills?  These are questions to ask when selecting a project.  The technology issue is one of does the library have the necessary equipment now, and down the road a year or two (or 20?) will the library be able to keep these collections online?  Are they in it for the long haul?

            As more and more libraries rush to digitize holdings just to be in the in-crowd, or as PR stunts, the institutions that spend the time and thought to carefully select their digital projects and make sure to document their decisions with posted selection criteria will stand out as truly excellent.  Hopefully libraries will begin to talk to one another, sharing their guidance.  Which is all it can be at this point, as Paul Ayris writes: “In truth, it can only be guidance rather than guidelines because relatively little is stable in a digital environment.”[1]  The technology is changing so quickly that standards wont stick – there is no sure footing.  But if enough libraries take the risk to create actual selection criteria, eventually, patterns will emerge and standards can begin to form.

Alexander Zimmerman Copyleft 02000