• creating minute-by-minute logs for all of the interviews in the series
• writing abstracts for each of the interviews
• choosing appropriate metadata fields, and cross-walking the local cataloging schema to the required DublinCore schema
• subject cataloging and description of the collection, series, and individual items
• entering all data into the batch upload spreadsheet provided by OhioLink, and running the macro action which created XML metadata records for each audio file
• burned all materials to DVD to be sent to OhioLink

Now, we will wait until they are published, which could take up to a of couple weeks depending on current workloads and priorities at OhioLink.  While it is possible to remotely upload items into the DRC, new users are limited to the default metadata fields, which were not sufficient in this case.  In order to use custom fields, we needed to physically give them the files.  Once they are published, those custom fields we used will be available for future projects.  Although this is certainly not an ideal situation, it is not such a big deal either.  I expect that it is a result of the DRC’s relative newness and that the process will be streamlined in the near future.

At this point, I am holding off on further comments about this project, which will be discussed in greater detail on the Project Report page of this site.

The plan is to take a series of 17 digital oral history interviews about the Shaker Lakes Nature Center (a subset of a rather large and diverse collection held by the History Department at Cleveland State University) and publish them to the Ohio Link Digital Resource Commons (DRC).  The challenge is to shape up the materials and documentation, and in the process, create a standardized and streamlined method for publishing the remainder of the collection (which consists of nearly 500 interviews in all).

At the present, I am working with scattered documentation for each of these interviews, so part of the process is to create a baseline of information that can be included across the collection.  On the most basic level, I need to be able to describe the actual interviews as discreet items (where are they located in all their formats and versions? what are the characteristics of those formats and versions? and so on…); next, I need to describe the context of the items (who contributed to the content and its creation and dissemination?  what can I tell the user about these contributors?  what is the series and how does it fit into the greater collection? what is the nature of the item? et cetera); finally, I must describe the intellectual content of the items (what is it about?).

Each of these tasks provide their own challenges – even when documentation is strong and consistent.  In this case however – due in part to the lower level of documentation, but also due to the nature of oral histories – the challenge is somewhat greater.  For example, with each of the interviews involved, I must listen to the audio files (ranging in length from 30 minutes to 2.5 hours), gain as much information as possible, and note topics of discussion in a minute by minute log.  These audio logs, which will be the basis for the abstract and table of contents for each item, will alone account for roughly 20 hours of work. That is just for 17 items, or about 4% of the total collection (providing a clear illustration of why these interviews, completed in 2006, have yet to be published in any form).

The Ohio Link DRC is built on Dspace and uses Dublin Core metadata fields.  Thus, I have started by establishing which metadata fields will be used, and which standards and controlled vocabularies will be employed to ensure consistency as this process continues (well beyond my practicum, and in all likelihood, with considerable work being done by non-LIS graduate assistants).  Once I have completed preparing all the metadata, I will transfer the fields into a DRC batch upload spreadsheet to upload the files and records onto Dspace.  I will have more to say about that process when the time comes.

Metadata fields for Shaker Lakes Nature Center oral history project: http://spreadsheets.google.com/ccc?key=p5XI6iRq54sEAWTOCmPN-fQ (Live Google Doc)

Metadata spreadsheet for items in Shaker Lakes Nature Center oral history project: http://spreadsheets.google.com/ccc?key=p5XI6iRq54sF8w7Uf4iEJdw (Live Google Doc)

Monday, 10am-6pm

Tuesday, 10am-6pm (with a 30 minute break to vote, of course…)

Wednesday, 10am-6pm

Thursday, 10am-6pm

Friday, 10am-6pm

Parrish and Bingham on Flickr

Parrish & Bingham on Cleveland Memory

This will be discussed further in my project report.

10am-2pm

• Rehousing: removed photos from album, put each photo in numbered archival-quality folders, put folders in archival storage container
• Digitizing: scanned and named the images according to original order of the album, created DVD storage copy containing archival copies (resolution: 600dpi, format: TIF, size: 100% ) and web copies (resolution: 72dpi, format: JPG, size: large side 600px)
• Cataloging: created collection in CONTENTdm, uploaded images, filled core and administrative metadata fields (including subject classification and creation of descriptive titles, etc. )
• Research: researched the subject in order to create accurate metadata and provide additional resources and introductory material for a future Parrish & Bingham website
• Other: when the images go active on Cleveland Memory, I will also create custom queries in CONTENTdm for use on the website.

Tomorrow I will be finished with my required 100 hours (and then some).  I plan to meet with Kif to discuss my final 50 hour project, which will consist of creating a web-accessible collection out of the 500 or so digital oral history files created by the CSU History Department.

10am-4pm

Here’s a rough draft:

Early History of Parrish and Bingham

Founded in 1894 as a trolley and bicycle frame manufacturer, the Parrish and Bingham Co. quickly grew to become an important component in Cleveland’s expanding industrial economy.  Located at the corner of Madison Avenue and West 106th Street – just across the street from the Cleveland auto manufacturer, Winton Motor Cars -  the Parrish and Bingham plant was ideally situated to enter the nascent automobile industry.  In 1923, the company merged with Detroit Pressed Steel Co. to become Midland Steel Products Company.  By the 1930s, with an expanded line of products that included steel automotive frames and axle housings for cars, trucks and busses; the West 106th Street plant became one of the central suppliers of automotive manufacturing components in the country, and “the largest producer of automotive frames.”

World War II and Beyond

Like many other manufacturers in the U.S., the West 106th Street plant rapidly adjusted to wartime production, turning out jeep and truck frames as well as parts for shells and tanks.  After the war, the company began to expand its interests beyond the automotive frames business.  Midland Steel Products Company became Midland-Ross Co. in 1957, beginning a process of diversification and acquisitions that would continue throughout the 1960s.  Industrial Rayon Corporation and the National Casting Company (both of Cleveland) are only two of the many national and regional manufacturers acquired by Midland-Ross between 1957 and 1969.

Beginning in 1969, the company began focusing its  research and development efforts on aerospace and electronics products and thermal processes, as well as steel castings, the latter of which dated back to the company’s 19th century origins as Parrish and Bingham Company.  Midland-Ross flourished during this period, as did the old plant on West 106th Street, which by the 1970s employed nearly 800  Clevelanders.  Despite corporate restructuring (leading to changing ownership) and periodic labor disputes, the plant managed to compete as one of only four automotive frame manufacturers remaining in the United States in 1976, and the nation’s largest manufacturer of frames for medium sized trucks.

Parrish & Bingham Images on Cleveland Memory

The Parrish and Bingham Collection at CSU’s Cleveland Memory Project consists of 80 photographs by Ernst-Eidman Photographers dating to (circa) 1920.  The photos were purchased by the library from a private collector, and were originally matted on linen pages housed in a single “screw ring” leather binder.  Most images are of the Parish & Bingham property at Madison Avenue and West 106th Street, including the factory floor and infirmary, as well as an interior mural and one aerial view facing south from Berea Road.  Assorted pieces of machinery; including cranes, presses, and blast furnaces; workspaces;  and warehouse and shipping and receiving areas are all well-represented.  Several images of men at work are also included.

10am-4pm

On a preservation note, we removed the album leafs from the binder prior to scanning (to avoid stressing the leather binding and linen pages) and decided to keep them separated after discovering some rust and jagged edges.

10am-5pm

When I’m done scanning, I will be entering the images and metadata into CONTENTdm for inclusion on website, as well as making archival/backup copies.

10am-5:15pm

1. Lewis Turco Collection – Finding Aid
2. Alberta Turner Collection – Finding Aid
3. James A. Miller Collection – Finding Aid

There are a few formatting issues which I have since corrected.  Hopefully, the corrections will appear online shortly.  We’ll have to wait and see if  there are any complications with re-submitting finding aids. Aside from the previous quirks I mentioned, I’ve also found that exporting MARC records seems to be problematic.  Although the output is readable and all relevant fields are automatically mapped, the record does not contain the leader (an essential component… it controls the MAchine Readable part of MARC).  There are also a few fields that have to be manually adjusted and/or added.  Here’s a sample output (.DOC file).

It’s been really interesting to learn the ins and outs of the EAD Creation tool, and to put into practice some of the archival standards I’m learning.

10am-4pm

Since I started my practicum, I’ve been very conscious of the time I’m spending on each task, trying to gauge what a typical professional might finish in a comparable amount of time.  Cataloging during my first week was very slow due in part to a fairly steep learning curve on some of the software, so it’s hard to judge how efficient I might be at that job if I had some more time to get used to the tools.  I think archival description is going a bit smoother, but I’m hanging now between two different modes of operation.  Part of me wants to get every last detail covered in processing this collection (including all the recommended preservation steps).  On the other hand, even a relatively small archive like CSU has a backlog of collections waiting to be processed, and spending too much time on one collection means another is that much further from seeing the light of day. (Not that I am responsible for such things, being an unpaid intern who’s here for only a few  weeks more, but I think it is useful to operate as if you have a stake in the outcomes).

One of the recent debates in the archival profession has centered around a 2005 paper by Greene & Meissner, entitled “More Product, Less Process” (aka MPLP), in which the authors call for less detailed and fastidious processing in order to save time and money, and to increase the number of publicly accessible collections.  On the other side of the argument are preservationists who feel that that cutting corners in processing will lead to problems down the road, as well as reference staff who feel that less detailed finding aids will lead to more time being spent helping users navigate a collection.  Some researchers also may have a problem with this approach, although, from what many sources state, they really don’t seem to care about finding aids, preservation or cleanliness, etc. — as long as they can see a list of what the collection contains and then find what they are looking for, they will be happy.  And isn’t that what it’s all about?

I highly recommend reading the original article:

More Product, Less Process [PDF]: http://www.midwestarchives.org/2006_Fall/readings/AA68.2.GreeneMeissner.pdf

…also, a few MPLP counterpoints:  http://www.archivesnext.com/?p=146

Thanks, Kif, for bringing this to my attention.

10am-4pm

The “archival mystery” I referred to in my previous post was in reference to an extensive series of manuscripts by another author, James A. Miller, whose works were found in the Turner collection, and whose identity could not readily be confirmed.  For a while, I considered the possibility that he was a prolific student of Turner’s; then, based on his apparent fixation on the Amish, I came to think he might be an “outsider artist” whom Turner had mentored; finally, I thought Miller was a Turner pseudonym.  As it turns out, he was a fairly successful poet in his own right and appears to have been a non-traditional (i.e. older) student at the Poetry Center at the time of his death in 1998.

After some queries and discussions, it was decided to break up the original order of the collection (no small decision in the archival field!) based on evidence suggesting that the donors had most likely mixed up two separate collections prior to bringing the materials to CSU Special Collections, or that in any case, the materials in question warranted their own designation.  Kif found the key evidence – a website once maintained by someone in CSU’s English Department, and quite possibly by Alberta Turner herself (view the website), which would explain in part how the two collections may have become commingled.

In any case, it has been an interesting project so far.  Tommorrow, I will go back to the EAD Creation Tool, consulting DACS and other standards, to finally finish the official finding aid.

10am-5:30pm

(with an hour break in the middle to go to a meeting for my “day job”)

I have come across a few “archival mysteries” in the process as well, which is actually kind of exciting.  We were unable to locate the original acquisition information, so when I came across what appears to be an anomalous series of items, I didn’t know what to make of it.  With Kif’s help, however, I think we figured it out and it appears (tentatively) that the donor may have mixed together two separate (but somewhat related) collections.  As a result, I may in fact be creating two finding aids.  Or alternately, we may find that the apparently anomalous items actually do belong together in some way.  Who knows.

More on this as I work throught he process in the coming days.

10am-4pm

• Inability to change unit type after creation.  If you make several units, encoding them as “files” and then for whatever reason, decide they are actually “items,” you are unable to change the description without deleting the unit and starting from scratch.  I suppose the solution would be to be very careful, but it would be nice if there was a way to simply edit the unit type.
• WYSIWYG rich text editor has some issues.  This was the biggest problem I encountered and it caused some headaches when it came time to validate the document.  While the editor gives you the option to create links and format fonts (italic, bold, etc.), it does not allow you to create valid XML links and formatting.  There is no way to look at the code of any given entry to check, edit, or verify the code being used.  What you end up with if you use the rich text editor is an invalid document.  Sadly, I had to go back through the document and unformat all entries, meaning no active links (just dead URLs) and no formatted text (no italics for titles, etc).
• Copy-Paste problems.  When you copy from another site, especially older sites it appears, you end up inheriting the HTLM code underneath the text – even though you cannot see this without exporting the XML file and viewing it with an external application.  Again, if you could look at the code, this would be easily fixed inside the EAD tool.
• The internal validator does not sufficiently point to where problems are within the document.  Again, the best solution is to use an external application.
• Inability to upload completed XML files.  This becomes a problem if you need to edit the document outside of the EAD Creation Tool and could also prevent the mass adoption that OhioLink is hoping for with its EAD Finding Aid Repository. Scratch that.  You can upload the EAD directly into the repository, just not in the Creation Tool interface.

Nevertheless, this is a fairly new tool and it is still extremely useful and robust in many ways.

Tommorrow, I begin working on an original finding aid, so I have some reading to do tonight.

10am-4pm

Using the EAD Finding Aid Creation Tool, hosted at our very own Kent State University, my task is to encode a “legacy” finding aid for the Lewis Turco Collection at CSU’s Special Collections for import into OhioLink’s Finding Aid Repository. Since the finding aid already exists, this is a fairly straightforward task – my contribution is ensuring that the finding aid is EAD/DACS-compliant, and free of encoding errors and other mistakes. We’ve also found some additional materials that will need to be added to the aid and encoded.

Later this week, I hope to get started on creating an original finding aid, which is a bit more involved and requires the analysis of acquisition records, etc.

10am-4pm

For the remainder of the day, I worked on more cataloging.  By the end of today, things seemed to be moving along nicely, so I’m pretty confident that a few more days would get me to a point where my work would be a lot quicker and a whole lot more reliable.  That being said, I am looking forward to moving on to EAD next week.

Me: The problem is not that I can’t do it.  It’s just that I’m very slow.  I could sit and work on one item for, like, 45 minutes.

Her: I think I don’t understand cataloging, because I can’t imagine how that would take 45 minutes.

Me:  I also think I don’t understand cataloging, for the same reason.

Her: Seriously, sometime you’ll have to explain what is involved in cataloging so I’ll understand.

Me: I can give you a concise explanation right now if you want?

Her: No, I’m driving.  I don’t want to fall asleep at the wheel.

It sounds like a bad syndicated comic strip, but it’s true.  This is the same girl who, eight weeks into my MLIS program, confessed that she pretty much stopped listening to my after-class banter around week two.  Direct quote: “You lost me at metadata.”

Despite my struggles with MARC/AACR2, I really like LCC subject cataloging in ClassWeb – I could do that all day.  I think what is most appealing about it is that you have to analyze the item (at least a little), and think about what it is, and then try to think like a user who is trying to find it.  It’s an interesting and challenging process, which takes a little imagination.  Good times.

All of these cataloging tasks really make you appreciate the amount of time and labor and money that go into running an effective library. Nevertheless, I will be glad to move forward to something new.  Next week, I will begin working with EAD finding aids, which will be entirely new to me.  More on EAD later.

10am-4pm

After the meeting, I sat in on an ebrary “webinar.”  Despite some technical glitches and some motivated salesmanship, it was fairly informative and quite interesting to see the business side of library technology.  I think CSU will be setting up an ebrary test run, so maybe I’ll get to poke around and see more of how it works (and/or doesn’t).

For the remainder of the day, I continued cataloging websites for the student subject portals.  I’m getting a little better at using the tools at my disposal, so now I just need to remember what I learned during those two semesters of MARC, AACR2, LCC, etc.

Looks like I have some homework tonight.

10am-4pm

All in all, a good start.

10am-4pm

It looks like I will be doing some cataloging (copy and original) so I am looking through some of the standards set by OCLC, OhioLink and CSU in preparation.

View: OCLC Bibliographic Formats and Standards

View: OhioLink Policy Handbook

View: OhioLink Database Management and Standards

Download [DOC]: Shen, Yuezeng. Procedures for Cataloging Electronic Resources. Cleveland State University, 2008

These articles should come in handy when evaluating Omeka, which the library has installed for testing.  For my real (i.e. paid) job, I recently visited the Center for History and New Media to discuss how Omeka can be integrated with CSU’s CONTENTdm-based collections to create exhibits that are more visually attractive, focused, and interactive than CONTENTdm usually allows.  However, at the moment, we are only capable of transmitting metadata using OAI-PMH, and have not yet developed an API capable of “scraping” actual objects (i.e. allowing Omeka to natively display objects that are located in a CONTENTdm repository).  So one of the things I will be doing for my practicum is investigating how Omeka can be used at the library – with or without the API.

It also appears that I will get to sit in on some meetings, such as the ebrary webinar mentioned in my last post, where technology choices will be discussed and software will be evaluated.

Based on the readings listed below, the major areas of consideration when selecting digital library software are:

• establish a clear and specific project statement that includes the scope of the project, infrastructure dimensions, and the overall purpose of the project
• explore the software functionality requirements that must be met to accomplish project goals and meet user needs
• explore practices at similar – and dissimilar – institutions to better understand the issues involved, compare a variety of approaches, and set a baseline for functionality
• determine the characteristics and practices of the user community using empirical data
• rank and prioritize functionality requirements to determine a baseline and to set forth a plan for development and added functionality over time if needed
• examine the viability of the vendor
• determine the security and authentication features of the software – including the granularity of user access and privileges – to determine how the software fits into existing institutional systems
• consider whether the budget and other resources available will meet the demands of the new software
• come up with a plan for training and documentation
• if the plan is part of a Universal Service Fund, it should additionally include an assessment of the institution’s  technology environment, a coherent strategy for meeting specific goals for improvement, and an established evaluation process for monitoring progress and making necessary changes mid-course

View: Connecticut State Library Checklist for Technology Plans

Download [PDF]: Cervone, H. Frank. “Some considerations when selecting digital library software.” OCLC Systems and Services: Int’l digital library perspectives, 22(2), 2006.

I am scheduled to attend several lectures and meetings next week.  Among them is an ebrary “webinar.”  ebrary is a platform for storing and distributing electronic resources such as  journals, e-books, and other documents, which may be acquired through purchase, subscription, or local creation (e.g. dissertations, training manuals, faculty publications, etc.).  I am looking forward to learning more about this and observing the evaluation process during and after the presentation.

Download [PDF]: ebrary Brochure

This work took about two hours (8pm-10pm).