| SURVEYS AND RESULTS
the course of the project, CineFiles staff designed surveys
to help us better understand the needs and practices of both
content providers and copyright holders regarding the use
of copyright-protected materials in the digital environment.
Below you will find links to pages containing results of our
survey and descriptions of our survey design and methodology,
as well as links to the surveys themselves.
Survey Design, Methodology, and Results
Survey of Content Providers
Survey of Copyright Holders
Providers Survey Form
Follow-Up Telephone Interview
Questions for Content Providers
Copyright Holders Survey (original)
Copyright Holders Survey (revised)
Content Providers Survey
of Content Providers
Survey Design and Methodology
In 2003, during the first phase of the project, we designed
a survey to help us collect data on the types of organizations
involved in image database management, descriptions of their
collections, and the ways in which they address copyright
issues. We consulted with experts in survey design and in
digital image databases to determine our list of questions.
We decided to create a relatively short survey, as we felt
this would produce the best result. Our final survey of content
providers includes seventeen
questions and was made available as an online document so
that responses could be easily compiled and tabulated in a
FileMaker Pro database for later analysis.
We created a separate FileMaker Pro database to collect the
names and e-mail addresses of institutions and individuals
we wished to survey. These included participants in workshops
such as “Copyright in the Digital Age”
and “School for Scanning”,
institutions that have received IMLS grants to develop digital
image databases, and database managers listed in the Clearinghouse
of Image Databases.
We e-mailed this group a request to complete our survey and
also posted the request to the mcn-l and Imagelib listservs.
From these efforts, we received forty-eight responses. We e-mailed
a second request two months later and received twelve additional
responses, for a total of sixty.
Complete survey results are available here.
As the results indicate, the great majority of respondents
were from nonprofit academic institutions, museums, and libraries,
involved with digitizing a wide range of materials. Among
the respondents, 93% work with large collections numbering
more than 50,000 items, with 28% planning digital image archives
numbering more than 100,000 items. However, most respondents
still had a lot of work ahead of them; ten respondents had
completed at least 98% of their planned digitization, forty-four had
completed less than 50%.
While thirty-one respondents work primarily with public domain materials
and thirty-one concentrate on those for which their organization holds
copyright, twenty-five work with materials under the copyright protection
of another party. (Respondents often checked more than one
response, creating totals that are higher than the number
of respondents.) Thirty-five of those working with copyright-protected
materials request permissions to make materials freely available,
while twenty-five pay licensing fees and five negotiate technological
strategies, such as password protection, to restrict use of
copyright-protected materials. Twenty-two respondents were
satisfied with the effectiveness and cost-efficiency of their
permissions clearance process, while twelve were not. Many respondents
commented that the high labor and administrative costs needed
to clear and track permissions discourages them from working
extensively with protected materials.
Follow-Up Telephone Interviews
Next, we identified survey respondents who work with copyright-protected
materials and who indicated their willingness to be contacted
with additional questions. We conducted eight follow-up telephone
interviews—five with art museum staff members and three
with staff from libraries. (Please see telephone interview
questions.) These interviews
were very informative, providing a wealth of detail on the
practices of individual institutions, and giving us useful
referrals to resources (many are included in the resource
list on this website).
Several of the phone interviewees mentioned the legal precedent
established in the Kelly v. Arriba Soft case (http://caselaw.lp.findlaw.com/data2/circs/9th/0055521p.pdf)
that allows the display of thumbnail images under fair use.
Many institutions digitizing art images therefore display
thumbnails to the general public without clearing rights,
and restrict full-size images to in-house use. Others discussed
the concept of risk assessment as “a big part of copyright
work.” A number of institutions display images with
ambiguous copyright status, but will, of course, take such
images down if they are contacted by the copyright owner.
In general, those who are grappling with securing permissions
agree that it is a time-consuming and labor-intensive process,
and therefore, they must allocate adequate time and resources
to imaging projects that involve working with
of Copyright Holders
Survey Design and Methodology
As we continued to collect data from content providers, we
also developed a survey for copyright holders. This survey
had two objectives; first, to learn more about each copyright
holder’s policies regarding granting permissions, and
second, to seek permission to display documents in CineFiles.
From past experience in rights clearance, we knew we would
get much better results by contacting copyright holders by
phone rather than in writing, so we formulated the questions
in the form of a phone script.
We soon realized that this approach—inquiring first
about policies and next requesting permission to display documents
in CineFiles—was less effective than we hoped, as the
copyright holders generally did not want to take the time
to answer survey questions. We therefore adjusted our methodology
to first concentrate on requesting permission to display materials
in CineFiles, and then asking policy questions later in the
conversation in a more indirect way, as demonstrated in the
revised script. This approach
proved much more fruitful and still allowed us to collect
significant data on copyright holder policies and practices.
Process and Workflow
As these interviews were conducted by phone and not online,
our copyright permissions analyst kept detailed written accounts
of her process and the copyright holders’ responses
so we could easily compile and analyze qualitative and quantitative
statistics on this aspect of the project. We tracked this
information in a Filemaker Pro database set up to manage digital
rights data and data specific to this grant project. In addition
to transcribing interviews and compiling statistical data
on type of organization, type of document, etc., she also
tracked the length of time taken to research current copyright
holders’ contact information, contact them and request
permission, conduct follow-up conversations or correspondence,
and obtain a final answer. Our process and workflow activities
can be found here.
In the course of our survey, we contacted 265 copyright holders
whose materials are represented in Pacific Film Archive’s
CineFiles database of film documentation. The group was made
up largely of for-profit organizations (43%) and freelance
writers (42%), with the rest being academic institutions (4%),
museums (4%), libraries (3%), other nonprofit entities (3%),
and government institutions (1%).
Generally, large organizations such as newspapers or magazines
have permissions departments that handle copyright clearance.
In some other cases an editor would handle our initial request
and subsequently hand it off to the legal department. In the
case of freelance writers, it was up to them individually
to grant or deny permission.
As we were working from documents indexed in CineFiles that
were primarily originally published in large newspapers and
journals, we usually started with these publishers. We were
often referred to the writers themselves, who in many cases
retained copyright, hence the large percentage of freelance
writers we contacted in the group described above.
Among the documents for which we sought clearance, 37% were
film reviews, followed in prevalence by other newspaper articles
(17%), book excerpts (14%), periodical articles (13%), film
showcase program notes (10%), film distribution publicity
materials (5%), and film festival program notes (4%).
Of the copyright holders, 16% surveyed had a written copyright
permission policy, and of those with a written policy, the
overwhelming majority were large newspapers or book publishers.
Regarding our question as to whether the institution retains
rights to its material or it reverts to the author, in the
case of virtually all of the newspapers and periodicals we
surveyed, the answer to the question was not simply one or
the other. In nearly all cases, there were "staff"
writers and "freelance" writers at a given newspaper
or periodical, and the rights to staff-written material were
retained while the rights to most freelance written material
reverted to the author (as upheld in Tasini
v. New York Times).
When asked about existing points of access for an organization's
material, 83% of the respondents had no infrastructure in
place for archiving or retrieving their copyrighted material.
Eight percent offered the option of allowing us to provide
links to archived articles on their website free of charge.
Four percent informed us that their material was available
at the site of an archiving service such as dowjones.com or
ProQuest (where there would be a fee for accessing the material).
Another 4% archived the material on their own website but
charged a fee for its access, and 1% had material on their
site that was password-protected and required a subscription
for access. Although no entity we interviewed contracted with
an online copyright clearance center to clear copyright for
their material, many individuals use such centers to seek
permissions to reproduce individual works in university course
readers, for example.
Use of Digital Rights Management Technology and Methods
Through this survey, we hoped to learn whether technological
restrictions, such as tethering (permitting documents to be
accessed for a specific length of time), providing read-only
access, or using other digital rights management methods,
would predispose more organizations to grant permission to
use their copyrighted materials. However, the organizations
or individuals we interviewed were by and large either too
big or too small to address this question in a meaningful
way. Most larger organizations had very specific policies
that generally did not address such technologies, while smaller
organizations and institutions were neither equipped nor interested
in such approaches. Approximately 1% of respondents were initially
receptive to the suggestion of our adding password protection
to our site, but in all cases, it ultimately made no difference
in their decision to grant or not grant copyright clearance
for our use. In many cases, however, where publishers do not
want us to display our digital scans of their documents, they
do encourage us to link to copies of articles on their websites, which can be done legally without specific permission.
Most organizations we surveyed still handle copyright permissions
on a case-by-case basis, which allows them a great deal of
flexibility and control in the use of their materials. Unfortunately,
however, this means that the effort to identify, locate, and
negotiate with copyright holders remains the burden of content
Our survey of copyright holders led us to conclude that new
digital rights management technologies were not currently
significantly impacting the use of these types of copyright-protected
materials in the digital environment. The rapid speed of technological
change and the major improvements in digital rights management
on the horizon may well spur future change. At present, however,
traditional copyright clearance methods, coupled with persistence,
still offer the best outcome.