Good afternoon, and welcome to the Joint RLG and NPO Preservation Conference, Guidelines for Digital Imaging. I am very pleased to be here, to welcome you, to inspire you, and to suggest some guiding principles for these deliberatio
ns.
- Don't reinvent the wheel -- or any of its spokes. Should we spend our time developing a set of RLG/NPO guidelines for digital imaging? And if so, why? As we all know, interest in things digital occupies the attention of many
people, some of whom have already given a lot of thought to suggesting guidelines for digital imaging. A number of the better ones have been produced by those assembled here. If we are to proceed in this direction, we must answer the
following questions: Is what is available inadequate? Incomplete? Inaccurate? Outdated? Obscured? Or even appropriate? However we decide to act, it will be our role to acknowledge this work, to take the very best from it, and to build
on it, not just replicate it. Our first task then should be to assess what's out there, where the gaps are, and what's the commitment from other bodies to create, update, and maintain important information.
- At least not an old wheel. If we do decide to move forward in creating a set of RLG/NPO guidelines for digital imaging, let's take care not to reinvent old wheels. We must clearly be mindful of what's come before, but also a
ssume a critical distance from it. We must be sensitive to differences not fully comprehendible today between things physical and things digital. It is natural to look to other image reformatting guidelines because they have served us
well and can serve us again, but only with an understanding that we're dealing with something altogether different. Consider the differences between microfilm and digital imagery. Microfilm represents a physical object, consisting of <
l>content fixed on a medium. In the case of digital imagery, an emerging definition of the digital object embodies both content and services. Unlike microfilm, the structuring, context, and provenance of a digi
tal file must be "explicitly captured and documented as it is created." (1) Or consider that in a microfilm version, contrast is fixed, but in a digital image it isn't. Digital images
are dynamic, where access and preservation requirements are more process oriented than with microfilm. In moving from analog to digital, there's a significant translation going on, and current models can not anticipate all of the resul
ts. We need to consider these compelling differences as we borrow from the old to meet the new.
- Acknowledge the full continuum -- As others have made clear, digital image conversion is intricately entwined with other functions, particularly those associated with management, preservation, and use. We must appreciate how
decisions made upstream at the point of capture can have ramifications all along the digitization chain, affecting functions and roles in unexpected ways. So if we seek to define guidelines for digital imaging, we must consider them i
n a very broad context. We must strive to understand the technical environment in which these materials will be accessed and used, as Columbia University has in providing technical recommendations for digital imaging projects conducted
by faculty, students, and staff. (2) And, in a networked world, we will be held accountable publicly for our decisions made; we need to look no further than the current trial of Clinton by the
American public to know how true this can be. (3) So we must think about how well significant informational content is conveyed, including what the Germans call the "testimonial" evide
nce that supports or clarifies the content through external formal features. But we must also consider how the file lends itself to processing (OCR, visual searching), to navigation, resource discovery, and versioning for access and us
e, to "dynamic updating with richer encoding," and to long term maintenance. (4) We know that we want to provide files of "enduring value" but we must also ensure that they co
nvey "endearing value." So we must concern ourselves with all manner of things technical: quality metrics, file formats, header information, structural metadata, the use, type, and level of compression, resolution, bit depth,
and so on. All of this would be hard enough in a static environment, but in a constantly changing one, it's more than a little mind-boggling. We must strive to go beyond meeting current service objectives to envisioning capabilities o
n the horizon-such as visual searching, vectorization, improvements in OCR, on the fly conversions. We need a lot of help from our friends.
- Beware of the hype. In an article on the continuing role of books, RLG's Walt Crawford warns against the "great technological hand wave." (5) He refers to a tendency to beli
eve that all things are technologically possible and within our reach, and to minimize shortcomings, placing faith in a belief that they will be overcome in "just a couple more years." How many of us have heard that the techn
ology just keeps getting cheaper and cheaper -- so that by tomorrow it will be free and presumably by the next day we'll be paid to use it? It is important to keep in mind that technology is not our friend. If things appear to be diffi
cult it's probably because they are difficult. While this seems common sense, we must also not be paralyzed by it, and we should beware the naysayers as well. Even some of our most forward thinkers can have their worldview impaired by
a lack of appreciation for change -- recall that in 1981 Bill Gates declared that "640K ought to be enough for everybody." Things do change, and often for the better. We must strike a delicate balance between wishing problems
away, and being too wedded to the way things work now. This conflict leads us to another principle.
- Own the technology, but don't let it own you. One of the key differences between microfilm and digital technology is the pace of change. With microfilm, we don't have to know that much about the specifics of the technology,
because we can have faith in its stability and its rules. But as consumers of digital technology, where there is no stability and the ground rules change all the time, we must assume a greater responsibility to understand what's here,
what's coming, what's transient, and to be able to see technological change in terms of library and archival interests.
Near term improvements in the post-processing of image information, for instance, may facilitate a shift in thinking about how to create the highest possible image quality for a given collection. A suggested new capture architecture
has the appropriate raw grayscale or color data collected from any scanner whose document handling capabilities suit the peculiarities of a particular item, such as a bound volume, a 35mm slide, or a 40 inch wide architectural drawing
. The scanner choice can be made on the basis of its physical suitability and the quality of its raw image data. All special processing from these various sources would then be performed in an off-line, largely scanner-independent mann
er. If this were possible, we would not be constrained by the variable and inconsistent processing offered within the many different scanners that are needed to overcome the physical peculiarities of each item in a collection. This wor
k would be particularly important in developing the means for capturing bound volumes without having to resort to disbinding, to endangering the volume, to creating photo-intermediates, or to a lowering of our quality requirements. (6)
We need to own the technology, yet not let it own us. This suggests to me that as we seek to define requirements, they should be as independent as possible from any specific technological approach. The constraints of today may not b
e there tomorrow, and we should avoid building an approach that becomes quickly outdated or superceded. A strong concern at Corbis, a commercial digital archive, is to be able to maintain the usability of digital photographic reproduct
ions on future unknown delivery and output systems. (7) As Paul Conway reminded us at RLG's first digital preservation conference five years ago, we should follow the advice of our colleague El
sie Freeman Finch. When we need a quarter inch hole, we should say so, rather than define our needs in terms of a particular drill bit.(8)
- Become a skeptic, especially about hard and fast guidelines. As noted earlier there is a plethora of information being presented as digital imaging guidelines, but a lot of it is contradictory, misleading, incomplete, or sel
f-serving. To paraphrase the New York Times, we need to follow all the advice that's fit to take. Consider, for instance, recommendations regarding resolution. Some say scan at the highest possible resolution, others recommend t
hat you avoid scanning images at very high resolution because it can cause an image to be blurred.(9) Several imaging "experts" argue that there is no sense in scanning color photogra
phic prints at resolutions above 200 dpi because there is no additional information to be captured. (10) It is true that a color print has far less information than the negative. But you stil
l have to look at it, see how big it is, and determine the level of significant detail contained in it. One can make a fairly decent scan at 200 dpi for most consumer variety color prints produced at the local drug store on smaller pri
nt sizes. But the same does not hold true for higher quality color prints generated through fine cameras and printed onto larger formats via a professional lab-or prints generated from transparencies above 35mm. As James Reilly put it,
"you can't be absolutist about any dictum that only talks about dpi for a whole (rather undefined) class of photos. (11) It's important to consider all the aspects governing resolution
choice with a color print or anything else.
- Come to terms with the notion of preservation in digital reformatting. It has been suggested that digital imaging can be used for preservation, for improved access, or for both. I believe this is not the case today. The prim
ary use of digital imaging into the near future will be to improve access. It may also be used for preservation and access, but not just for preservation purposes alone. If only preservation purposes are to be served, other refo
rmatting options to ensure continued viability of the informational content are probably better choices to make. We'll hear later in the conference an update on digital preservation efforts, but we all know that we don't decreasethe preservation problem by relying on digital information, we only increase it. As Terry Kuny put it, "being digital means being ephemeral."(12)
A German report on the intrinsic value of archives and library material explicitly rejects the use of digital imaging for preservation purposes because of its lack of fixity, but does acknowledge its use for improving access. (13) In fact, there is renewed interest in a hybrid approach that marries microfilm for preservation and digital imaging for access. (14) In the Unit
ed States, where the brittle books program opened the door to a definition of preservation as the permanence of information documents, there has even been a resurgent demand for retaining the physical objects as well. (15) In some American universities, digital imaging is used to replace originals when an analog paper or film copy is created, and the University of Michigan is experimenting with digitization and on-line acce
ss only. A growing number of governmental agencies and businesses are turning to the technology to conduct office back file conversions, with or without an analog backup, and with or without official archival sanction. With few excepti
ons, though, the primary motivation has been to improve access, even when preservation is also a concern.
So if we are to assume digital imaging is primarily motivated by access considerations, where does that leave the preservation side? We should acknowledge that the notion of preservation is relative, and turn our attention to creati
ng digital objects that we are prepared to keep indefinitely (however long or short that might be). Even if we are not thinking long term retention, immediate applications and current uses require that we create a fairly rich file or s
et of files, hence the need to focus on the requisite image quality and metadata for creating "access masters." These criteria should be based on how well the files lend themselves for processing and for presentation in a var
iety of formats and on a range of devices and media. At this point, if we think preservation digital master at all, we should be judging it by whether the file can produce preservation quality analog backups -- preferably COM -- and ho
w to preserve the accompanying requisite metadata in an eye-readable form. (16)
I like the Australian assumption that resources should be devoted to retaining digital materials "only for as long as they are judged to have continuing value and significance." (17)
We should strive to create access master files in a way that makes them worthy candidates for long term retention-so that disposition decisions are based on continuing value and functionality, not limited by technical decisions m
ade at the point of conversion or anywhere else along the digitization chain. The role of those concerned with preservation should encompass the creation of image files that are "preservation worthy." Peter Lyman and Brewster
Kahle distinguish between "born" digital and "borne-again" digital -- files created by converting from other media. (18) Once in digital form, these distinctions may not b
e that significant, at least for digital imagery. I would argue we just don't want to create "still born" digital files -- that we need to create files that are "fit for purpose" as our UK colleagues would say, and
cost-effective over their full life cycle. Neil Beagrie and Daniel Greenstein of the Arts and Humanities Data Service argue in their strategic policy framework for creating and preserving digital collections that "how data is crea
ted will impinge directly upon how it can be managed, used and preserved at any future date." They suggest that we need case studies to "demonstrate the benefits of any additional investment toward long-term preservation duri
ng data creation in terms of efficiencies and use later in the life cycle of the resource."(19)
- Presume Web or Web-like delivery. Deanna Marcum, President of the Council on Library and Information Resources in the US once said, "The notion of a stand-alone digital library seems anathema to its intent." (20) Digitization is a process of separating information from physicality of form and place -- and by implication digital copying and digital distribution, which today and into the foreseeable futur
e equates to network access. With digital conversion, cultural institutions become concerned about distributed resources and distributed users. This leads us to consider quality and utility at the point of use much more than we ever ha
d to face in the analog world. Just as we are concerned about quality at the start of the process, so must we be concerned at its end. We must evaluate the effects of technical choices made in creation and presentation of digita
l image surrogates for a range of document types.
Although there is much anecdotal evidence, and a number of studies conducted involving digital image use, there has been no systematic assessment of the cumulative effects of technological choices on the networked display of digital
image material.(21) In 1990, Michael Ester published the results of an influential Getty study on users' assessments of digital image quality of works of art, based on such factors as resolu
tion and bit depth.(22) Subsequent studies have noted that the quality of digital masters will directly affect the quality, utility, and expense associated with creating derivative images for
on-screen display, printing, and image processing. (23) A number of institutions, the Library of Congress and the National Archives amongst them, have promulgated guidelines for the creation
of access images for Web display. (24) The National Archives and Corbis, Inc. have both gone so far as to create electronic grayscale targets designed to assist users in calibrating their moni
tors to optimize the presentation of delivered digital files. (25) In addition, there are some helpful sites, including the YaleC/AIM Web Style Guide, that provide useful recommendations on ch
oosing file formats and optimizing graphics. (26) A number of companies and institutions have also produced Web-accessible reports or graphic examples of the effects on image quality of specif
ic technical choices, such as the type and level of compression used. (27)
Despite their valuable contributions to the field, these resources have largely ignored how the range of technical choices-file formats, compression processes, scripting routines, transfer protocols, web browsers, and the like -- ca
n affect perceived image quality at a networked workstation. So too must we consider the lag in technology adoption at the user's end. For instance, the Museum Education Site License Project, co-sponsored by the Getty Institute, involv
ed the creation of fine access images of paintings, photographs, and three-dimensional objects. These images met many of the researchers' needs and expectations at high end user workstations, but were almost useless in their largely po
sterized form when received on an older machine with a VGA monitor incapable of supporting more than 256 levels of color.
As cultural institutions increasingly turn to the web to make retrospective resources accessible to a broader public, issues associated with image quality, utility, and delivery at the user's end get raised. Creating high quality di
gital image files is not an absolute guarantee that this will happen. The preservation community must become involved in decisions made at the point of image manipulation, scaling, compression, and other technical matters in order to s
afeguard the user's rights to access satisfactory digital information.
- Acknowledge the gap between the ideal and the possible. Having said all that has come before, I do think that we have to exercise judgement and do the best we can, given various resource constraints and the technological lim
itations of the here and now. Something can be conceptually strong, but just not work. Consider for instance the case of the Library of Congress, which went out with a Request for Proposals (a.k.a. an Invitation To Tender) for the imag
ing of graphic materials with a stated preference for image files that were 12 bits per channel. The staff discovered that it was difficult to store and use these files, and have used them instead to create, as does Corbis, good qualit
y 8-bit per channel images -- but they do not retain the larger bit files.(28)
There is also an emerging debate within the library imaging community in the US over the use of compression in the digital master. Some argue that any compression is bad, but others argue that tough choices have to be made-do we sav
e fewer items by making our requirements so demanding? The authors of the previously cited Strategic Policy Framework for Creating and Preserving Digital Collections concluded:
- "The standards and best practices which promised to facilitate and reduce the cost of a data resource's long-term preservation were not always those which promised best to facilitate and reduce the cost of its intended use.
The standards and best practices, which promised to ensure a data resource's maximum fitness for purpose were also not always affordable or technically achievable. Accordingly, the selection of standards and best practice frequently i
nvolved a range of compromises between data creation aims and costs."(29)
We will often need to make decisions that conflict with our quality requirements on the one hand or our financial capabilities on the other-this is to be expected, but we should be making those choices in an informed manner.
We need more information on lessons learned about when things go wrong (people are more than happy to tell you when they go right!). A good example of this is found in the TASI report, Creating Digital Image Archives. TASI vi
sited several digital imaging library projects to identify what worked and what did not, and to determine whether current standards could be practically implemented. One particular problem, experienced by the project team of the Higher
Education Library for Image eXchange (HELIX), arose when trying to buy scanners: "the advice given by suppliers and bureaus was technically wrong. For example, one supplier stated firmly that a scanner could be PC driven, when in
fact it could only be run from a Mac." This, together with other technical problems led to some delays within the project.(30)
- Undertake the doable. In a presentation to the 1992 annual meeting of the Research Libraries Group, Don Waters suggested a number of enabling principles with which to view the challenges of imaging technology. Among them wer
e the KISS (Keep it Simple, Stupid) principle and the adoption of an incremental approach. (31) These two points are still valid today. We should strive to undertake what's doable, from the pe
rspective of our collective energies and knowledge base. Consider for instance the case of the International Color Consortium, which undertook several years ago the challenge of solving all the issues associated with color management.
The task was too daunting -- at least in the very near term, and the ICC strategy has now shifted to an incremental approach, addressing what's doable now and developing building blocks for an iterative effort. If we become too ambitio
us, we may end up further behind than if we had attempted nothing at all. I can't stress enough the need for us to succeed in our deliberations. Let's plan for it.
- Consider guidance rather than guidelines. All these principles lead
me astray from my appointed rounds to suggest that we consider guidance instead
of guidelines. I suggest this for a number of reasons:
- One size doesn't fit all. In a world with multiple stakeholders
and multiple perspectives it becomes difficult to create guidelines that
suit all circumstances. Institutional purposes vary, so it seems problematic
to create hard and fast answers, to standardize technical choices that can
scale across all institutions, all collections, and all use objectives.
The imaging guidelines proposed by both the National Archives and the Library
of Congress are simi lar in many ways, but differ in significant ways, based
on conclusions experts at each of these institutions reached regarding their
institutional mission, their collections, their users, and current technological
constraints. In other words there is no one right way to do things that
is applicable in all circumstances.
- Reasonable, well informed people can disagree. We must acknowledge
that differences of opinion in an uncertain world are a healthy sign. We
need to experiment with a range of approaches and evaluate their impact
over the lon g haul. But if experts can disagree on such fundamental principles
as compression and bit depth, and the use of enhancements at the point of
capture, how do we sort out the "truth?"
- Digital imaging is a process not a prescription.I am more of a
believer in an approach than essential truths. Digital imaging is not so
much a recipe as a series of choices that has to be made, in which we must
balance compe ting requirements along the way. Guidelines by their very
nature are static or restricted to the specifics of the environment in which
they were produced, whereas guidance looks at the way things flow and adjusts
accordingly. As we uti lize this technology, we want to make informed decisions,
to understand the consequences of choices made upstream to what flows thereafter,
and to be prepared to right our course occasionally as things change.
- But people want simple answers, not a process. And they will interpret
guidelines more broadly than they should. Take for instance the recommendations
coming out of Cornell for the use of 600 dpi bitonal scanning. This recom
mendation is based on an examination of a specific body of material (brittle
books containing text and simple line art) that exhibits remarkably uniform
characteristics typical of the commercial printing processes of the 19th
and early 20th century. It does not extend to office documents, to other
publications, to manuscripts, or to graphic and photographic materials --
yet time and time again, I hear of institutions requiring 600 dpi for their
imaging projects that have nothing to do with brittle books, because that
is the Cornell "standard." Six hundred dpi is not for all things,
and may not be required in a program where access alone is the driving force.
- It is often difficult to state requirements in terms of precise objective
characteristics.Sometimes what we want is for the image to look right,
and that can't always be quantified or characterized in terms of precise
requir ements. If we rely too closely on technical definitions, we may end
up being too wedded to the numbers and sacrifice quality to consistency.
Our goal is not to reproduce the grayscale bar but to convey the essence
of the original objec t. (32) Or
it may not be that at all. Photo historians refer to the concept of "rendering
intent" in reformatting photographs -- what the photographer meant
to convey, rather than wh at is recorded on the photographic medium, which
calls for an interpretive assessment. This, I would argue leads back to
stating requirements in terms of what we want, rather than prescribing how
to get there. Let me not be misundersto od. For some things we can and should
take a definitive stand-we should require objective measurements, but understand
them as indicators rather than as guaranteesof image quality.
In photographs, tone reproduction is the most import attribute of image
quality: "If you get your neutrals right, most of your color management
work is done."(33) We should
require the use of targets, but understand that i n relying on an evenly
distributed grayscale, the image will look darker and muddier on screen
and in print. The final decision on image quality should be made by a person
sitting in front of a monitor, who decides what looks "goo d."
As Supreme Court Justice Potter Stewart reputedly once said about pornography,
"I can't define it, but I know it when I see it."(34)
- Our knowledge is not evenly developed. We know more about some
kinds of scanning than we do others, and we might be close in reaching agreement
on best practices for some categories of materials. For other materials,
special ists in their fields suggest that we are far from consensus on how
best to achieve digitization. (35) And
for some materials, the most that can be said may be that we know what we
don't want. Given this situation, we can not be uniformly definitive in
stating requirements for digital imaging. But we can be knowledgeable about
the tradeoffs associated with different courses of action.
I can't help but wonder if we would not be better served by assuming the
role of Sherpa rather than chef. If you will forgive the mixed metaphors,
one can navigate with any number of sea charts. But the chances of arriving
safely ar e greater if there is someone along who has made this trip many
times, and knows how to avoid the shoals, depending upon the prevailing winds,
the vessel used, and the behavior of the gods. Where experts disagree or institutional
missi ons conflict, our role may be the most beneficial if we can help institutions
interpret and then synthesize the differences of opinion, and the circumstances
under which one approach may be better than another. In some areas, we should
be prepared to challenge assumptions, or to offer a "Good Housekeeping
Seal of Approval" for a particularly well reasoned methodological approach.
We should also point out where solid work remains to be done, and alert other
s via a "watch this space" to initiatives underway that may shed
some light on particularly thorny issues. Why, for instance, should we spend
the next three days agreeing on guidelines that include imaging pictorial
materials if the Visual Arts Data Service draft guidelines for "Creating
digital information for the visual arts" offer us all we need?(36)
And, if they don't, then the Council on Library and Information
Resources is also preparing to undertake this task. In other words, let's
provide guidance rather than guidelines. Let's develop a product that allows
us to continue to retain relevancy as the technology and our understand ing
of its uses change, and as others develop solid guidelines for specific aspects
of this work. I believe that RLG/NPO are in an ideal position to vet for us
the work performed by others: to examine, investigate, and evaluate it in
a thorough manner -- via a process in which we can have a great deal of confidence
and trust.
So how does guidance work?
It is pretty easy to suggest that we focus on guidance rather than guidelines,
but just how do we go about this? We know how to do guidelines -- we have
a pretty good track record in this area -- and the process itself represents
a comfortable, well-trodden path. Developing guidance is more challenging,
representing a paradigm shift in what we do, and making this shift will be
challenging. I do, however, believe that we have no choice. Our heart may
be in creatin g guidelines, but our head should be with guidance. We will
ultimately fail in the attempt to create generalizable guidelines because
guidelines are contextual-they are specific to institutional settings, and
may not scale beyond those constraints or to other environments. As Sabinne
Susstrunk from Corbis points out, "choosing the correct scan resolution
is dependent on the purpose of the archive," not on some absolute. (37)
Even though the temptation is strong to extrapolate from one situation
to another, the reality is that what works for the Library of Congress probably
will not work for a local historical society.
Guidance, on the other hand, can be developed to enable an institution to
step through a process with the end result leading to the creation of their
own specific set of guidelines. Guidelines should be a by-product of this
process, not the starting point. Deciding on a course of action, what to do
and what not to do, should be based on the reasons for digitizing, the nature
of the source documents, the institutional mission, available resources, the
technical in frastructure, and users requirements and capabilities. Guidance
begins with understanding the context, provides a process for data gathering
and decision-making, points to available guidelines and a means for assessing
their applicabil ity, and finally leads to the development of guidelines that
are specific to institutional needs. Guidance should lay out a uniform process
in which prejudices are stated up front, decisions are well defined and documented,
and evaluat ion is built in all along the way. Guidance is an inductive process
that enables institutions to specify, create, and implement their own guidelines,
based on the specifics of their environment instead of relying on a pre-defined,
limi ted set of options. One such model may be represented in the work of
the Science Museum in the UK. The Museum is in the process of creating a digital
image policy that includes "cost-benefit analysis of storage, access,
and image quality options to guide staff in selecting appropriate standards
and formats for image content which is being considered for digitization."(38)
As we break for the next few days into separate groups, we should think about
providing a model process for guidance to be used by institutions to develop
guidelines for digital imaging. We should think about the impressive amount
o f work already accomplished, and consider where agreement on basic requirements
can be reached, for instance in the area of metadata. We should acknowledge
where there is divergence of opinion based on solid reasoning, and strive
to pu t this in the context of decisions made rather than universal truths.
We should recognize that preservation may be used in some instances and not
in others, but that long-term utility should be a guiding force. We should
consider the b est "bang for the buck" in what we come out with,
and come out with something doable -- the world is watching. We should focus
on a product that will best meet the needs of those of us assembled here today,
because if we can meet our needs, I guarantee we will go along way in meeting
the needs of others.
References:
- Neil Beagrie and Daniel Greenstein, "A Strategic
Policy Framework for Creating and Preserving Digital Collections," Arts
and Humanities Data Service, Version 4.0, July 14, 1998, p 9. http://ahds.ac.uk/manage/framework.htm back
to text
- Columbia University, "Technical Recommendations
for Digital Imaging Projects,"
http://www.columbia.edu/acis/dl/imagespec.html back t o
text
- Nancy Gwinn made the point that "everything libraries
do in the digital world will be more visible to more people" in " Mapping
the Intersection of Selection and Funding," Selecting Library and Archive
Collec tions for Digital Reformatting, Research Libraries Group, 1996, p.
58. back to text
- Beagrie and Greenstein, "A Strategic Policy Framework,"
p. 26 back to text
- Walt Crawford, "Paper persists: why physical Library
collections still matter" ONLINE, January 1998,
http://www.onlineinc.com/onlinemag/OL19 98/crawford1.html back
to text
- Anne R. Kenney, Louis H. Sharpe II, and Barbara
Berger, "Illustrated Book Study: Digital Conversion Requirements of Illustrations,"
in Research and Advance Technology for Digital Libraries: Proceedings of
the Second European Conference on Digital Libraries, Heraklion, Crete, Greece,
September 21-23, 1998, edited by Christos Nikolaou and Constantine Stephanidis
(Springer, 1998), p. 279-294. back to text
- Sabine Susstrunk, "Image Production Systems at
Corbis Corporation," RLG DigiNews, Vol. 2, No 4 (August 15, 1998),
http://www.rlg.org/preserv/diginews/ back to text
- Anne R. Kenney and Paul Conway, "General Address:
From Analog to Digital: Extending the Preservation Tool Kit," Digital Imaging
Technology for Preservation. Proceedings from an RLG Symposium held March
17 a nd 18, 1994, Cornell University, Ithaca, NY, edited by Nancy E. Elkington,
Research Libraries Group, Inc, 1994, p. 12. back to text
- Jane Williams and Alan Locke, "Image Capture &
Manipulation: Guidelines and Procedures," in Building and Delivering Digital
Image Archives, TASI,
htt p://www.tasi.ac.uk/building/image_cap1.html; Rice University Advanced
Visualization Lab, "Understanding Dots Per Inch (DPI)." back
to text
- Wayne Fulton, "A Few Scanning Tips;" Ibid., Rice
University Advanced Visualization Lab, "Understanding Dots Per Inch (DPI)."
back to text
- James Reilly to Anne R. Kenney, email, Sept. 9,
1998; also Don Williams to Anne R. Kenney, email, August 24, 1998. back
to text
- Terry Kuny, "The Digital Dark Ages? Challenges
in the Preservation of Electronic Information," International Preservation
News, No. 17, May 1998, p. 13. back to text
- Angelika Menne-Haritz and Nils Brubach, "The Intrinsic
Value of Archive and Library Material. List of criteria for imaging and
textual conversion for preservation. Results of a DFG project." http://www.uni-marburg.de/archivschule/intrinsengl.html
back to text
- Stephen Chapman, Paul Conway, and Anne R. Kenney,
Digital Imaging and Preservation Microfilm: the Future of the Hybrid
Approach for the Preservation of Brittle Books, Council on Library and
Informati on Resources,
http://www.clir.org/programs/cpa/hybridintro.html#full back
to text
- A review of disposition policies for microfilmed
materials is being conducted at a number of institutions, most notably The
New York Public Library, which came under criticism for discarding microfilmed
mate rials. See, Mark Singer, "Missed Opportunities Dept.: Did the New York
Public Library let some history slip through its fingers?" in "The Talk
of the Town," The New Yorker, January 12, 1998, p. 29-30. Currently, New
York State Assembly man Richard Brodsky is pursuing state legislation to
require libraries receiving any state funding to deposit copies of de-accessioning
policies with the State. back to text
- This definition is advanced in the hybrid paper
mentioned in footnote 14. back to text
- National Library of Australia, "Statement of Principles
for the Preservation of and Long-Term Access to Australian Digital Objects,
April 1996(
http://www.nla.gov.au/preserve/digital/princ.html) back
to text
- Peter Lyman and Brewster Kahle, "Archiving Digital
Cultural Artifacts, Organizing an Agenda for Action," D-Lib Magazine, July/August
1998, http://www.d
lib.org/dlib/july98/07lyman.html back to text
- Neil Beagrie and Daniel Greenstein, "A Strategic
Policy Framework for Creating and Preserving Digital Collections," Arts
and Humanities Data Service, Version 4.0, July 14, 1998, p. 65, and p 6,
http://ahds.ac.uk/manage/framework.htm
back to text
- Comment by Deanna Marcum at a SOLINET conference
on digitizing projects for libraries and archives, "To Scan or not to Scan:
What are the Questions?" May 2, 1996, Atlanta, Georgia. back
to tex t
- Armitage, Linda H. and Enser, Peter G.B., "Analysis
of User Need in Image Archives." Journal of Information Science, vol. 23,
no. 4, 1997, pp. 287-299; and Keistler, L.H. "User Types and Queries: Impact
on Image Access Systems," in R. Fidel et al. (Eds.) Challenges in Indexing
Electronic Text and Images. Medford, NJ: Learned Information Inc., 1994,
pp. 7-22. back to text
- Ester, Michael, "Image Quality and Viewer Perception,"
LEONARDO, Digital Image-Digital Cinema, supplemental issue (Pergamon, 1990).
back to text
- Stephen Chapman and Anne R. Kenney, "Digital Conversion
of Library Research Materials: A Case for Full Informational Capture," D-Lib
Magazine, October 1996, http://www.dlib.org/dlib/october96/cornell/10chapman.html.
Michael Ester also makes this point in a number of publications, most recently
in Digital Image Collections: Issues and Practice, Commission on Preserva
tion and Access, December 1996. See also Charles S. Rhyne, Computer Images
for Research, Teaching, and Publication in Art History and Related Disciplines,
Commission on Preservation and Access, January 1996; Roger S. Bagnall, Digital
Imaging of Papyri: A Report to the Commission on Preservation and Access,
September 1995; Janet Gertz, Oversize Color Images Project, 1994-1995. (Online
access or ordering information on publications of the Commission on Preservation
and Access is available at:
http://www.clir.org/pubs/reports/reports.html.) Final Report of Phase
I. Commission on Preservation and Access, August 1995; and Carl Fleischhauer
and Ricky L. Erway, "Reproduction-Quality Issues in a Digital-Library System:
Observations of the Reproduction of Various Library and Archival Material
Formats for Access and Preservation," an American Memory White Paper, Library
of C ongress, December 1992.
http://memory.loc.gov/ammem/ftpfiles.html back
to text
- See Library of Congress recommendations in "Digital
Formats for Content Reproductions," by Carl Fleischhauer, at
http://memory.loc.gov/ammem/formats.html and those of Columbia University
at http://www.columbia.edu/acis/dl/imagespec.html.
The National Archives and Records Administration has produced "Guidelines
for Digitiz ing Archival Materials for Electronic Access," which are available
on the Web at
http://www.nara.gov/nara/vision/eap/eapspec.html back
to text
- http://www.corbis.com/
back to text
-
YaleC/AIM Web Style Guide,
http://mirrored.ukoln.ac.uk/web-authoring/caim/caim/contents.html back
to text
- See for example, the following sites
http://www2.ncsu.edu/bae/people/faculty/walker/hotlist/graphics.html
(which covers graphics viewers, editors, and utilities); "Image Compression
Using Human Visual System Models, Wavelet Transform Coding, and Massively
Parallelizable Algorithms,"
http://esd.gsfc.nasa.gov/ESS/annual.reports/ess9 5contents/app.gci.mathews.html;
"Technical Overview of Cartesian Perceptual Compression,"
http://www.cartesianinc.com/Tech/tech-overview.html; "LC Maps: Scanning
Ca rtographic Materials Panoramic Maps,"
http://lcweb2.loc.gov/ammem/pmhtml/pandigit.html; and "Understanding
image file formats,"
http://www.cobb.com/tma/9508/tma89501.htm back to text
- "The Library of Congress Requests Proposals for
Digital Images of Pictorial Materials, National Digital Library Program,"
LC Contracts and Logistics Services, LCRFP 97-9, see C.4.2, "Desirable Image
Types," p. C-6 (available online at:
http://memory.loc.gov/ammem/ftpfiles.html); Carl Fleischhauer, "Digital
Formats for Content Reproductions," Library of Congress, July 13, 1998,
http://memory.loc.gov/ammem/formats.html;
Sabine Susstrunk, "Imaging Production Systems at Corbis Corporation," RLG
DigiNews, August 15, 1998; Franziska Frey, "Digital Imaging for Pho tographic
Collections: Foundations for Technical Standards," RLG DigiNews, December
15, 1997, http://www.rlg.org/preserv/diginews/.
back to text
- "A Strategic Policy Framework for Creating and
Preserving Digital Collections," Arts and Humanities Data Service, p. 23
back to text
- Jane Williams , "Project Management Summary: HELIX,"
in Building and Delivering Digital Image Archives, TASI,
http://www.tasi.ac.uk/building/helix4.htm l back to text
- Don Waters, "Electronic Technologies and Preservation
," Commission on Preservation and Access, publication No. 25,
http://www.clir.org/pubs/reports/ reports.html. back to
text
- Quote from James Reilly in his lecture on digitizing
photographs, Cornell Digital Imaging for Libraries and Archives Workshop,
July 22, 1998. back to text
- Quote from Don Brown in Anne R. Kenney and Oya
Y. Rieger, Using Kodak Photo CD Technology for Preservation and Access:
A Guide for Librarians, Archivists, and Curators, Cornell University Library,
Ithaca, NY , 1998, p. 14. On-line version available at:
http://www.library.cornell.edu/preservation/pub.htm back
to text
- Quoted in James M. O'Toole, "On the Idea of Permanence,"
American Archivist, Vol. 52, No. 1 (Winter 1989), p. 22. back
to text
- See for instance, David Yehling Allen, "Creating
and Distributing High Resolution Cartographic Images," RLG DigiNews, Volume
2, No. 4, August 15, 1998,
http:// www.rlg.org/preserv/diginews/. back to text
- Catherine Grout, Janine Rymer, Jane Williams,
Karla Youngs, "Creating Digital Resources for the Visual Arts: Standards
and Good Practices," Visual Arts Data Service Guides to Good Practice, Guide
I (draft, 1998) back to text
- Sabine Susstrunk, "Image Production Systems at
Corbis Corporation," RLG DigiNews, Vol. 2, No 4 (August 15, 1998),
http://www.rlg.org/preserv/diginews/ back to text
- Neil Beagrie and Daniel Greenstein, "A Strategic
Policy Framework for Creating and Preserving Digital Collections," Arts
and Humanities Data Service, Version 4.0, July 14, 1998, p. 24. back
to text
![[RLG logo]](file806.gif)
![[JISC logo]](file699.gif)
![[UKOLN logo]](file209.gif)
![[NPO logo]](file667.gif)