EAD Implementation Project � Final Report by axUtrWCN

VIEWS: 0 PAGES: 10

									Table of Contents


 Executive Summary ...................................................................................... 2
 Project Outcomes .......................................................................................... 2
   1.     Advisory Committee ............................................................................ 2
   2.     Staff training & use of encoding software ............................................ 3
   3.     Units’ plans for the EAD Project .......................................................... 4
   4.     Critical mass of finding aids from all ASC units .................................. 4
   5.     Centralized point of access for all finding aids. ................................... 5
 Detailed Comments and Observations ........................................................... 5
   General Comments ..................................................................................... 5
   Additional Outcomes .................................................................................. 7
   Unexpected Outcome ................................................................................. 8
 Future of EAD in ASC.................................................................................... 8
 Appendices: ................................................................................................. 10
   A.     EAD Project Charter .............................................................................
   B.     Special Projects Request Form: .............................................................
   C.     Project Expenditures .............................................................................
   D.     EAD Training for Students ....................................................................
   E.     “Using XMetaL to Create EAD Finding Aids” .........................................
   F.     Detailed Discussions by Unit ................................................................
   G.     Statistics: ASC Totals............................................................................
   H.     Statistics: Collection-Level Records .......................................................
   I.     Statistics: By Unit .................................................................................
                                                 University of Minnesota. Archives and Special Collections
                                                                EAD Implementation Project – Final Report
                                                                                           June 17, 2005



Executive Summary
       The EAD 1 Implementation Project started in March 2004 and was
funded with $100,00 from the Libraries Leadership Committee. It developed
from the EAD Planning Project that was completed in late 2003.2 The
implementation project built upon the Implementation Recommendations from
the planning project and used that project’s Best Practices and Application
Guidelines as its foundation.
       The Implementation Project was directed by Leslie Czechowski. As part of
the project, a Web page was created that would provide information about the
project as it proceded. Included are links to documents from the Planning
Project, links to ASC units and their finding aids, links to EAD help pages at
other institutions, monthly project reports, and templates, stylesheets and
other resources. [http://wiki.lib.umn.edu/Staff/FindingAidsInEAD]
       The Implementation Project had three main parts: to train ASC staff to
implement EAD as part of their regular descriptive practice, to encode legacy
finding aids, and to select and implement a delivery mechanism. The project
has been extremely successful. Project staff have recorded 1650 hours and have
created 575 finding aids. Additional funding in January 2005 allowed us to hire
a part-time library assistant to create collection-level finding aids that has
provided an additional 600 finding aids. In addition, staff in the ASC units have
created still more finding aids so that we now have over 2500 finding aids
available for researchers on the Web. The success of the project is due to the
hard work and commitment of all staff in the Archives and Special Collections
Department.
       The EAD Project Charter (see Appendix A) listed six outcomes (described
below in five sections) for the project. Information about them is detailed below.
Following that are details of additional information regarding what we learned
during the project, unexpected outcomes, and the future of EAD in ASC, along
with extensive appendices.



Project Outcomes
1. Advisory Committee
    From the Project Charter: Establishment of an advisory committee to oversee the
    implementation of EAD, address policy issues that come up, and facilitate liaisons between
    the project archivist and relevant staff in SC/Ar, MAC, DLDL, and DCU.

      An Advisory Committee was created at the beginning of the project that
included one representative from each ASC unit. Members are:

1 Encoded Archival Description [EAD] is the new standard in the archival community. It
is an XML-based standard that makes it possible to provide access to archival finding
aids (used for description of collections) in a platform-independent, electronic format
structured to facilitate search and discovery.
2 Members of the initial EAD Task Force were Lara Friedman-Shedlov (Chair), John

Butler, Leslie Czechowski, Christine DeZelar-Tiedman, Carrie Seib.


                                                                                                        2
                                                  University of Minnesota. Archives and Special Collections
                                                                 EAD Implementation Project – Final Report
                                                                                            June 17, 2005

       Linnea Anderson, Social Welfare History Archives
       John Barneson, Children’s Literature Research Collections
       Barb Bezat, Manuscripts Division
       John Butler, Digital Library Development Lab
       Leslie Czechowski, Chair
       Karla Davis, Special Collections & Rare Books
       Christine DeZelar-Tiedman, Technical Services
       Lara Friedman-Shedlov, Kautz Family YMCA Archives
       Karen Klinkenberg, University Archives
       Daniel Necas, Immigration History Research Center
       Carrie Seib, Charles Babbage Institute
 The committee met on an irregular basis to facilitate communication between
unit staff, make decisions about implementation issues and encoding practices,
and give input about the delivery system. A number of staff gave presentations
about special encoding practices that they had developed (see details in later
section).


2. Staff training & use of encoding software
    From the Project Charter: Training, appropriate to varying staff needs, as outlined in the
    planning project's implementation recommendations. Integration into routine finding aid
    creation of encoding software tools (purchased or developed in-house), as outlined in the
    planning project's implementation recommendations.

        The EAD Task Force had recommended that all staff in ASC that created
finding aids and were responsible for description should learn EAD. At this
point the units still operated independently, and this decision was made based
on the “lone-arranger” model in the archival profession. 3 However, levels of
expertise with EAD varied greatly. Our first step, then, was to provide training
for all staff that wanted or needed it. Czechowski provided a brief “EAD Lite”
presentation for curators and other staff giving an overview of EAD without
specifics of encoding.
        Following recommendations of the EAD Task Force, we purchased 18
copies of the xml-encoding software, XMetaL, for all unit staff and for two
reading room computers to be used by students. There were two training
sessions for the staff (led by Czechowski and Friedman-Shedlov); each lasted
two hours. We gave a detailed overview of EAD and XMetaL and then gave staff
time to practice encoding a small finding aid, adapting and using the minimum
standards template created by the Task Force. Staff were shown how to
personalize the template for their own unit by adding “boilerplate” information
that would be the same for all their unit’s finding aids.
        Two students were hired to work on encoding, focusing on the more
routine data entry for the detailed contents of collections. Much of their work
was simply re-keying data found in finding aids that did not exist in an
electronic format. The first two students were not trained to fully encode an
EAD finding aid; we used their time for data entry. But the last student hired

3I.e., a single archivist handles all activities within a repository from accession to
description to reference.


                                                                                                         3
                                                  University of Minnesota. Archives and Special Collections
                                                                 EAD Implementation Project – Final Report
                                                                                            June 17, 2005

was trained to understand the entire EAD document and to encode more than
the detailed description. She will continue to work on EAD encoding after the
project ends and can provide more knowledgeable assistance to unit staff. The
training given to the students has been documented (see Appendices D & E) so
that unit staff can train students and other staff in the future.
       Staff in six ASC units currently use XMetaL to encode parts of their new
finding aids in EAD. [See section below for alternatives to using XMetaL for
encoding]. Two units use a database approach to finding aid creation and follow
different procedures for encoding. Staff in the James Ford Bell Library and the
Tretter Collection (SCRB) have relied on the expertise of EAD project staff to
encode their finding aids.
       As the project went on, Czechowski provided assistance to staff as
needed, helping them to manuever their way through the intricacies of EAD and
XMetaL. After a short while, other staff had developed their skills to the level
that they served as coaches and mentors for each other—an ideal situation that
will prove invaluable after the formal end of the project.


3. Units’ plans for the EAD Project
    From the Project Charter: A plan in place for each unit, including processes and routines, for
    encoding new and legacy finding aids on an ongoing basis.

        At the beginning of the project, Czechowski met with each unit in
Andersen Library to make a plan for encoding, especially their legacy finding
aids (in both paper and electronic formats) and to develop processes and
routines. The plans varied, but some common approaches were used.
        Database approach. CLRC and IHRC entered data about their
collections into databases. Project staff, along with volunteers, did data entry
for each unit. Staff in the two units can output EAD finding aids and have done
so for a large percentage of their collections.
        Large finding aid approach. A number of units used the project staff—
students especially—to get their largest finding aids into electronic format. For
example project staff spent 59 hours encoding a finding aid from SWHA and 98
hours on entering data for a YMCA finding aid.
        Most used and/or most important collections’ approach. Other units
had project staff work on their most important collections first, or the
collections that researchers used most often. [Which isn’t to say that the units
focusing on the Large finding aid approach weren’t having their most important
collections put into EAD.] For example, the Manuscripts Division had the
project staff encode the John Berryman collection early in the project.


4. Critical mass of finding aids from all ASC units
    From the Project Charter: A critical mass of encoded finding aids representing collections
    from all SC/Ar units.

       There have been no reports in the literature from other such projects
about how long it takes to encode finding aids. Initially we were hesitant to
state how many finding aids would be created during our project. However, we


                                                                                                         4
                                                  University of Minnesota. Archives and Special Collections
                                                                 EAD Implementation Project – Final Report
                                                                                            June 17, 2005

believe we have been very successful in creating a larger “critical mass” than
any of us expected. By mid June 2005 there are 2500 EAD finding aids
available for the public to use. This does not include completed finding aids
from IHRC and CLRC that haven’t yet been added to the publicly-available
group. During June, additional finding aids will be created.
        To put this number in perspective, there are 725 finding aids on the
Bentley Library site, and the Five College Archives & Manuscript Collections
project has 900 finding from Amherst, Hampshire, Mount Holyoke and Smith
Colleges and the University of Massachusetts Amherst.
        Reporting accurate statistics about the number of finding aids currently
available (beyond the numbers of finding aids created during the project) and
what percentage of each unit’s descriptions they represent is a difficult task. In
round figures, most units have well over 50% of their finding aids and
collections’ described in EAD and available over the Web. Two units (CLRC and
SWHA) have 90-95% of their collections described in EAD finding aids.


5. Centralized point of access for all finding aids.
    From the Project Charter: A centralized point of access, via the World Wide Web, for
    encoded finding aids from all units. This tool, will be developed in cooperation with the
    Digital Library Development Lab in accordance with the specifications outlined in the
    planning project's implementation recommendations.

       This is one area in which the project did not reach its goal. A small group
was charged with investigation of software for deliverying finding aids. The EAD
Tech group included four ASC staff (Barneson, Czechowski, Friedman-Shelov,
Necas) and two staff from the DLDL (Butler, Bill Tantzen). We decided that
purchase of an existing product was preferrable to building our own delivery
system and that DLXS-XPAT was the best available software. The purchase of
the product, begun in the winter 2004-2005, has just been finalized. The EAD
Tech Group has begun conceptual development of web pages for the finding
aids with Jen Tantzen’s guidance.




Detailed Comments and Observations
General Comments
       Some ASC units have existed for over fifty years; others are more recent.
Because the units were geographically and administratively separate, they
operated as independent units. Before the existence of the Internet, archives
developed various strategies for describing their collections, and those at the
University of Minnesota were no different. Because the collections differed,
description differed. The Northwest Architectural Archives created databases
describing each item in a set of drawings. CLRC had an extensive card catalog
with one card for each author or illustrator’s book and the types of materials in
the collection listed. University Archives had typewritten narrative descriptions
about collections, some with extensive subject lists.


                                                                                                         5
                                            University of Minnesota. Archives and Special Collections
                                                           EAD Implementation Project – Final Report
                                                                                      June 17, 2005

        The EAD Project led to a change in descriptive practices in many units.
Some units took the opportunity to revise older finding aids when they were
transferred into EAD. SWHA, for example, had legal-length folders listed at the
end of their older, typewritten finding aids. In the electronic environment we
could insert the legal folders where they belonged with regard to intellectual
content and series’ assignment rather than by physical location. Additionally,
accretions to collections could be entered into the finding by their intellectual
content rather than just as an appendage to the initial finding aid.
    We found that divergent descriptive practices among units called for a
uniform model for finding aids. The use of our Minimum Requirements
Template, based on discussions among curators with the Special Collections
cataloger even before the Task Force began their work, served as a content
model as well as the format for the EAD finding aid. Some of the ASC units did
not enter into our project intending to reengineer finding aids, but soon
discovered that the activity of converting finding aids into EAD resulted in
better descriptive practices, especially for units with finding aids created forty
years ago (when processes were different).
    University Archives, for example, had some small finding aids in which the
detailed contents were written in a descriptive manner and confusingly
embedded in the narrative regarding the scope and content of the collection
rather than clearly listed in one section. The Manuscripts Division, on the other
hand, had an inventory or calendar of materials for each collection with
accession information in one set of files and biographical information about
collection creators in other files. The EAD project gave them the opportunity to
bring together all this information in one document. In other units (SWHA for
example) accretions to collections had been inventoried separately because the
finding aid was typewritten. This project gave unit staff the opportunity to
intellectually arrange collections appropriately while retaining the physical
arrangement. In most units finding aids were written in different styles
depending on the age of the finding aid and the supervising archivist. This
project gave ASC staff the opportunity to unify descriptive practice within their
unit and to more closely align them with other units.
    Quantifying time spent encoding EAD finding aids has been a significant
part of the project, but reporting averages is meaningless given the different
types of legacy finding aids.
         University Archives has paper finding aids with simple contents lists
           that were re-typed into EAD in 3-5 pages (in the initial finding aid) per
           hour.
         SWHA has more hierarchically complicated paper finding aids in
           which data from 1-2 pages could be entered into EAD per hour.
         These ranges are consistent for a finding aid in electronic format in
           which project staff cut-and-pasted data from the original into EAD.
         For SWHA finding aids (in paper) we encoded 1.8 linear feet per hour.
         For CBI finding aids (mostly in Word or html) we spent 1-4 hours per
           cubic foot in creating the EAD finding aid (3 hours average).
         Creating a collection-level record requires only ¼- ½ an hour when
           all data is readily available. If the archivist has to do research in order
           to create the collection-level document, ¾ to one hour is required.



                                                                                                   6
                                           University of Minnesota. Archives and Special Collections
                                                          EAD Implementation Project – Final Report
                                                                                     June 17, 2005

          If contents’ details are in a well-formatted Access database, it takes
           less than 15 minutes to transfer data into EAD
    Unit staff spent varying amounts of time proofing, revising, and editing
finding aids created by project staff. Smaller finding aids (2-3 pages) required
only 10-15 minutes of review. Staff from YMCA, SWHA, and University Archives
spent an hour or more on some larger finding aids and those that required
revision or the addition of more details.
    As the project developed, staff developed new ways of doing things which
they shared with their colleagues. The innovative environment fostered by the
project has given many staff the opportunity to develop creative problem-
solving.
        Seib shared with her colleagues ways to manipulate data in Word
           documents (software in which many older finding aids were created)
           to automatically transfer them into EAD.
        Czechowski and Bezat took instructions for automating transfer of
           data from Access into EAD and developed a process that would work
           for more complex collections.
        Building on that process, Klaassen, L. Anderson, and a student
           assistant developed an even more elaborate procedure to transfer
           data from finding aids with multiple component levels and ordering
           them in specific ways, using Excel, Access, and EAD. This process is
           best suited for hierarchically-detailed descriptions.
    We expected that we would need the cooperation of and collaboration from
other Libraries staff, notably DLDL, Technical Services, and LEO; that proved
to be the case. Christine Dezelar-Tiedeman and John Butler worked closely with
the ASC group from the inception of the project. Michael Johnson created a
macro to convert MARC records to EAD. John Chapman and Betsy Friesen are
currently working to automate conversion of EAD to MARC.XML which will
reduce the time spent to create MARC records for each collection from the
finding aid.


Additional Outcomes
    As ways of encoding evolved, especially with the use of automated processes,
we developed a general sense of best procedures of EAD encoding (both new
and legacy finding aids). Many units now will create new contents lists in
Access, Excel, or Word tables since it is a faster and more efficient method of
inputing data, especially if we use student labor. It’s easier for students to learn
those programs (and many students already use them), than to learn EAD.
    Collection-level finding aids: better than nothing. Nine months into the
project it became clear that, although we were making excellent progress
creating EAD finding aids, we would not be able to complete a finding aid for
every collection. We requested and were granted additional internal funds to
hire an assistant for the project for six months [see Special Projects Request,
Appendix B] and hired Karen Spilman who had EAD experience. The focus of
this project was to create collection-level, EAD finding aids for collections that
did not have any type of description in electronic form (html, MARC, etc.). These
finding aids include brief narratives in the scope and content and biographical/



                                                                                                  7
                                           University of Minnesota. Archives and Special Collections
                                                          EAD Implementation Project – Final Report
                                                                                     June 17, 2005

historical notes plus other elements required by our Best Practices. They do not
include a contents list. These collection-level finding aids are similar to the
MARC to EAD finding aids that were created early in the project for SWHA and
IHRC. The finding aids may be enhanced in the future—or may not—depending
on the unit and the importance and use of the collection. We believe that it is
better to have basic information about many collections available online rather
than complete finding aids for only a few collections. This additional part of the
project gave us the opportunity to provide electronic access to a great majority
of our collections (except for some units’ unprocessed collections or collections
for which description is still in process).
    We have learned that it takes about ¾ hr. to create the average collection-
level record when data has to be found in a variety of locations (accession files,
biographical files, etc.). Spilman created 600 finding aids from January through
the middle of June for the following units.

                   CLRC                          100
                   Manuscripts                   247
                   SWHA                           52
                   U Archives                    178
                   SCRB                           23



Unexpected Outcome
       The most unexpected result of this project was the organizational
change that developed because of it. Staff in all the units worked together—
sharing expertise, developing new processes, and planning and managing a
project for the benefit of all. All staff—curators, archivists, and staff—could see
the value of cooperation in a project that served the entire department. Our
hope is that this level of cooperation in a positive and successful project will
lead to future cooperation, collaboration, and departmental decision making for
the benefit of all. It may prove to be a valuable first step as our nine units move
from independent entities into a whole that is greater than its parts.


Future of EAD in ASC
        Christine DeZelar-Tiedman will manage the transfer of finding aids into a
location for public access. She will work with unit staff to create MARC records
for finding aids that have been created during the project. Most units have not
had MARC records created, so this will entail a great deal of time. We hope that
the EAD to MARC conversion will facilitate this process.
        The EAD Advisory Group will continue to meet on an occasional basis,
especially as the Tech Group gears up with the DLXS part of the project. The
DLDL is commited to bringing up the delivery and search mechanism for
finding aids as are the ASC staff.
        The descriptive climate in ASC has changed since the beginning of the
project. Staff realize the value of EAD-encoded finding aids and, for the most


                                                                                                  8
                                          University of Minnesota. Archives and Special Collections
                                                         EAD Implementation Project – Final Report
                                                                                    June 17, 2005

part, will continue encoding new finding aids in EAD. As time, staff, and
funding allows, they hope to continue to reformat legacy finding aids into EAD.
For some unit staff, however, maintaining (or developing) EAD expertise will be
difficult. If a person only creates a few finding aids per year, s/he will not be
able to remember all the complicated details of EAD. In the beginning, the
original Task Force envisioned the process as if each unit was independent—
there needed to be at least one person per unit with the ability to encode finding
aids in EAD. But because the environment has changed in ASC, some staff
think that it might be more functional to have one person who has EAD
responsibilities for all ASC units. Unit staff can create finding aids in a variety
of software (Word, Access, Excel) and the EAD staff member could facilitate
transfer of the data into EAD.




                                                                                                 9
                                     University of Minnesota. Archives and Special Collections
                                                    EAD Implementation Project – Final Report
                                                                               June 17, 2005



Appendices:
       A. EAD Project Charter

       B. Special Projects Request Form:
          “Improving Descriptive Access to Special Collections: Encoding
          Legacy Finding Aids”

       C. Project Expenditures

       D. EAD Training for Students

       E. “Using XMetaL to Create EAD Finding Aids”

       F. Detailed Discussions by Unit

       G. Statistics: ASC Totals

       H. Statistics: Collection-Level Records

       I. Statistics: By Unit




                                                                                          10

								
To top