The Archivists’ Toolkit

...now browsing by category

 

We’re back! Bootcamp, processing, and progress so far…

Friday, April 4th, 2014
Training_Processing

New project team during minimal processing bootcamp at the University of Pennsylvania.

Hello again! Time has flown by, and we’re just getting the blog started again by recapping the current PACSCL/CLIR Hidden Collections Processing Project of 2013-2014. I assumed responsibilities of Project Manager in August 2013 and it’s been a whirlwind of activity from the very first day. I had to quickly assess and plan how we would minimally process 46 collections containing materials from the 18th to 21st centuries, all specifically related to Philadelphia history. Processing will require us to process at a rate of 4 hours per linear foot at 16 different repositories over the course of one year.  In addition to 12 veteran participating repositories, we welcome four new institutions to the project, including two non-PACSCL members. With this project, we hope to refine, confirm, and better establish guidelines for applying minimal processing to a wide range of collections and types of institutions and creating high-quality finding aids for our ever-expanding collaborative site.

Surveying at Germantown Historical Society.

Surveying at Germantown Historical Society.

As you may recall, this project builds upon the predecessor processing project lead by Holly Mengel and Courtney Smerz from 2009 to 2011. Having served as one of the processors on that project, I began my work as Project Manager already very familiar with the “PACSCL” methods and approaches established by the first team. My familiarity with these approaches, along with additional archives management experience, gave me a bit of a running start, but I immediately found that I have my work cut out for me. More about the challenges and lessons I’ve learned so far will be chronicled in later posts.

In August, I quickly got started by surveying the collections selected for the grant that had not been surveyed previously by the fabulous PACSCL Survey Initiative Project. I followed and expanded upon the guidelines already previously established in earlier projects to assess these new collections. In September and October, I was able to assemble a fabulous project team of six processors and one assistant, who all attended the bootcamp training week designed to establish a good overview of the PACSCL approaches to minimal processing and the Archivists’ Toolkit. After training, I assigned pairs of processors to our first three repositories (Temple University, the Philadelphia Museum of Art, and the Union League of Philadelphia) to kick off the year’s-worth of processing work ahead of us.

First day of processing at Temple University.

First day of processing at Temple University.

Already with many challenges and successes along the way that will be detailed further in the coming weeks on the blog, we hit our six-month mark this week right on track! At our halfway point in the project come mid-April, we will have processed an approximate total of 762 linear feet for 22 collections in 9 repositories, at an average rate of 3.45 hours per linear foot. Please stay tuned as we continue to add more frequent updates about our progress, lessons learned, and interesting finds!

Excel to EAD-XML to AT—the spreadsheet from heaven.

Monday, March 19th, 2012

Unknown size: small.

Although it seems like a million years, it actually was not so long ago that our students were processing at the Independence Seaport Museum.  While we were there, we were faced with one of the limitations of our minimal processing time frames.  The archivist there, Matt Herbison (now at Drexel University College of Medicine Legacy Center) had a few spreadsheets detailing information on ships’ plans—information that made the collections truly useful to researchers.  Problem was, there were thousands of entries in the spreadsheets and we knew that our processors could never re-key or copy/paste that information into the Archivists’ Toolkit in the time allotted for the processing.

Because we knew that this information would really make a difference for users, we thought and thought of ways to make this work, but our best solution involved saving the spreadsheet as a pdf and linking to it from the finding aid–not very elegant. And then Matt, who really is extraordinarily techie, created this amazing spreadsheet that solved the problem.  To sweeten the deal even more, he offered Courtney and me the use of the spreadsheet for the project.

I will now make a very bold statement:  this spreadsheet made it possible for us to finish the project within the time frame.  Not only did we use it at the Seaport, our processors used it for original data entry at repositories that had spotty internet connections, technical troubles, and/or did not adopt the Archivists’ Toolkit.  Our Archivists’ Toolkit cataloger used it as a starting point for almost all electronic legacy finding aids.

Matt has offered to share this spreadsheet with everyone.  It is available here and we have created a guide for using the spreadsheet.  In a nutshell, each column in the spreadsheet maps to specific field in the Archivists’ Toolkit.  It has three levels of hierarchy below the collection level, so it not the tool of choice if your finding aids has sub-sub series and items, but for most modern finding aids, it is the ticket.  I should say, though, that it is not necessarily a quick process if you are starting with existing data … time needs to be taken to combine columns, format data, and check for errors.  If you know how to use regular expressions, you can really streamline some of this work.  If you are doing original data entry, the use of the spreadsheet is incredibly efficient for getting container lists into the Archivists’ Toolkit.

This means that anyone with knowledge of MS Excel can create finding aids and take legacy information from an electronic format to xml.  Pretty awesome! I will say that a little knowledge of EAD is very useful and understanding the Archivists’ Toolkit will make decisions in data entry easier.  Many of our students preferred working with the spreadsheet rather than the Archivists’ Toolkit, but it is a matter of preference.  I think it is a little harder to see the hierarchy when using the spreadsheet, but it is a thousand times easier fix error in Excel than in the Archivists’ Toolkit.  Check it out, try it out and see if it changes your life.

Yes, I did say that … I think it could change your life!

Thanks SO much to Matt Herbison!

Legacy finding aids: a trial (by any definition)!

Monday, February 13th, 2012

Unknown size: small.


77 “substandard” or legacy guides are now in the Archivists’ Toolkit and final editing is underway.  And I am happy about that … however, almost none of these look as good as they could or should.  Garrett Boos, Archivists’ Toolkit cataloger, and I spoke many times about the limitations of this part of the project.

We decided that there were several problems:  working remotely from the collections; the format, structure and quality of the finding aids that were given to us; and, to be perfectly honest, our own expectations for the final product.

Before Garrett started, I decided that working remotely was going to be the most logical way to approach this part of the project.  Garrett worked in our office at Penn and entered the collections into our own instance of the Archivists’ Toolkit.  We then exported the finding aids from his AT and  imported them into each repository’s instance of the Archivists’ Toolkit.  I decided to have Garrett work at Penn primarily because of logistics—otherwise, he would have had to work at 18 different repositories and, as we have learned, technology and space are two of the greatest challenges of the project.  Not to mention the instances when security clearances would need to be run, etc.  However, now that Garrett is done with the project, I have been trying to decide if it would have been better for him to work on-site and I am torn.  On the one hand, it would have made a lot of factors easier—especially checking on locations, vague titles and missing dates, to name only a few.  On the other hand, it would almost certainly have stopped being a “legacy finding aid conversion” project and turned into a “reprocessing” project. So I guess I need to stand by my decision to work off-site, even it was limiting.

Unknown size: small.

The reason I say that it would have turned into a “reprocessing” project is because Garrett and I think that at least 60% of the collections should have had some physical and intellectual work before the finding aid was considered final.  As with all aspects of this project, the legacy finding aid component was an experiment and therefore, the grant allowed repositories to send us any “substandard finding aids.” This resulted in several types of “tools.”  Garrett took them all on:  lists, card catalogs, databases and more traditional finding aids.  The biggest problem we found was that very few of these guides were organized hierarchically which meant that we had to do a lot of guessing—was something a folder, or was it an item?  Should the paragraph connected to a folder title be added as a scope note or was it actually part of the folder title?  What to do with the information about the contents of a letter, or the condition of the material?  What happens when there is no biographical/historical note and no scope and content note?  Thank goodness for email and helpful repository staff! 

I should say that there were a number of finding aids that came to us in absolute perfect shape … putting that finding aid into the Archivists’ Toolkit was a piece of cake and the resulting finding aid was beautiful. Others that were written before finding aids were standardized did not work nearly so well. Because we forced non-hierarchical guides into AT, a system designed to organize information hierarchically, some of the finding aids are actually less user-friendly than the originals. Many of these legacy guides had item level description, something our stylesheet doesn’t handle well, resulting in what Garrett and I have termed, “really ugly finding aids.” Moreover, of 77 finding aids, only 15 did not require some enhancement of biographical/historical or scope and contents notes–which is pretty tricky when working off-site. Titles and dates almost always needed to be reformatted for DACs compliance. Our primary goal was to maintain every bit of information that was in the original, but it worries me that we have created online guides that are potentially overwhelming and off-putting to researchers.

Some repositories have told me that I should not worry—that getting the guide online is enough.  Others, though, I know are really disappointed with the result. We surveyed our participating repositories about the effectiveness of the project and their satisfaction, and while we have not heard from all, the component of the project that proved least satisfying is the legacy finding aid component. I know that it is, by far, the part of the project with which I am least pleased.

Does this mean that you should not do a legacy finding aid conversion project?  No!  Do a legacy finding aid conversion, but do it with some structure and guidelines!  In order to have a successful legacy finding aid conversion project, we learned that repository staff will have to do some (or alot of) front line work prior to unleashing the guide on the cataloger.

Before handing over a finding aid, repository staff should identify (in pencil is okay):

• Folder title (underlined in one color)
• Folder date (underlined in another color)
• Box number
• Folder number
• If there is additional material, into what field in the Archivists’ Toolkit/EAD should it be entered?
• Biographical/historical note (does not need to be narrative, but the information should be provided by an “expert”)
• Scope and content note (same as the bio note)

If, as you go through this process, it becomes obvious that reprocessing is necessary, take the collection off your conversion list and place it on a priority list for processing.  Processing the collection may be quick and speedy and your result will almost certainly be better! In fact, I think, in some cases, we spent more time forcing data into AT than it would have taken to reprocess the collection.

Identifying these essentials should result in finding aids that are more standardized and allow researchers greater access to your awesome stuff. Don’t count on it being a quick process, however: the prep work is time consuming, the conversion is time consuming, and the proofing and editing is REALLY time consuming. This is not a task that can be placed only on the person converting the finding aid … even after the finding aid was in AT, Courtney and I, with fresh pairs of eyes, found lots of mistakes in spelling, hierarchy and grammar which would have been embarrassing and, even worse, would have potentially prevented people from finding that for which they were looking. Which is, of course, the whole point of all our work!

More Product Less Process: Embracing flexibility in finding aids at Drexel University Archives

Thursday, August 4th, 2011

Written by Rob Sieczkiewicz, Archivist, Drexel University Archives

Drexel University Archives and Special Collections was one of the first repositories to participate in the PACSCL Hidden Collections project; processors came to Drexel in Fall 2009. As a result we have changed our procedures for publishing finding aids.  In the spirit of the “More Product, Less Process,” our goal is to provide access to collections as quickly as possible, with minimal concern for pretty finding aids.  We had been using Archivists’ Toolkit for almost 18 months before the PACSCL project processors arrived, and were enthusiasts.  Before AT, creating and publishing finding aids was a laborious process, with not so pretty results.  AT allowed us to export EAD easily, using a stylesheet created by the American Philosophical Society, slightly modified with Drexel information.  After a few months, we decided we to revise the spreadsheet to match our website, which basically meant that one of our staff of two had to teach herself how to edit EXtensible Stylesheet Language (XSL); this took a while, but the result was lovely.  At the same time, inspired by the UMarmot catablog created by Rob Cox at UMass Amherst, we moved our collection descriptions to WordPress, relieving us of the chores of HTML editing was WordPress.  The new platform and finding aid stylesheet looked great and worked just fine… until we needed to move away from WordPress and onto the same Content Management System (CMS) at the rest of the Drexel Libraries: Drupal.  Moving to Drupal broke our stylesheet leaving us with a lovely-looking Drupal web site and unreadable finding aids.  However, revising the style sheet to match that new site would have been a taken quite a bit of time.  Asking whether such an effort would be worthwhile, we determined that if the default Archivists’ Toolkit stylesheet was good enough for the PACSCL project, it was good enough for us.  Greene and Meissner say to invest your limited resources wisely – for us, the wisest investment was to put up the most basic finding aids, with minimal customization or adornment, and then process more collections, do more outreach, create more exhibitions rather than build the perfect XSL stylesheet.

We also upload finding aids to the PACSCL finding aids site.  This process is slightly redundant, however, and requires a bit of HTML editing.  Would it be a wiser use of resources to eliminate this redundancy?  Why put finding aids in two places?  We could simply link from the Drexel Archives web page to the PACSCL finding aids site.  For some repositories, such as those who lack access to an institutional web page (or simply lack a web page), this is the only option.  For others, giving up control of display is unthinkable.  But for some repositories, like Drexel University Archives, it presents yet another option to consider in the quest to provide the most access to our patrons by making wisest use of our limited resources.

Efficiencies and Access at Haverford College Quaker & Special Collections

Monday, August 1st, 2011

Written by John Anderies, Head of Special Collections, Haverford College

Unknown size: small.

Haverford College Quaker & Special Collections was one of the first institutions to be treated to the excellent work of Holly, Courtney and the fabulous student processors (hi, Forrest and Leslie!) hired for the Hidden Collections project. As a semi-official Guinea Pig, we really benefited from the extra time and attention given us by the PACSCL processing team.  All involved did first-rate work and brought some much needed order to 10 of the high-research-value collections in our backlog.  Participating in the project also jumpstarted our adoption of Archivists Toolkit to process new collections, has inspired us to find additional ways to open our holdings to researchers, and has provided our staff with ample opportunities to debate the pros and cons of minimal processing!

Today, we now record all accessions and process all new collections in Archivists Toolkit.

For accessions we record all gifts no matter the format (manuscripts, archives, books, photography and fine art) and any purchases that are not reflected in the acquisitions module of our ILS (such as manuscripts and photography).  Eventually we hope to include retrospective accessions in AT too.  In addition to the original 10 finding aids produced by PACSCL, we have completed 19 more in AT, all of which now reside on the PACSCL EAD Repository hosted at Penn, in addition to our local web server.

Our instance of Archivists Toolkit is installed on a Tri-College server located at Bryn Mawr College and serves the needs of four individual repositories across the consortia: Bryn Mawr Special Collections, Haverford Quaker & Special Collections, Friends Historical Library at Swarthmore College, and the Peace Collection at Swarthmore College.  Accessions and Resource (or collection) records for our four repositories are partitioned within AT.  However, we do share the tables for Subjects and People, which is very useful when the topics of our collections overlap, which they frequently do.

In addition to moving ahead on creating new finding aids in AT, we have spent the past year making our legacy finding aids more accessible.  Previous efforts at moving our finding aids into the 20th century had produced only a handful of fully searchable guides online and a mish-mash of Word files, PDFs, XML files, Excel files, ASCII text files, and Filemaker Pro databases living on a single staff computer, inaccessible to our researchers without the direct intervention of staff.  A decision to “not let the perfect be the enemy of the good” finally freed us from our paralysis and has produced extraordinary results.

Unknown size: small.

When the PACSCL crew left us in 2009 we had—in addition to their 10 finding aids created in AT—approximately 45 other finding aids online. By agreeing that it was better to supply our researchers with something “quick and dirty” than nothing at all and through the dedication of our students and staff, we turned all of the other finding aid formats into PDFs and mounted them on our web server.  These are listed on two web pages in both Collection Name and Collection Number order and the complete lot of nearly 250 finding aids is searchable using a Google Custom Search.  The results lists are not always pretty and neither are some of the finding aids, but for the first time the majority of our materials are discoverable online and our researchers seem pleased with the access.

As the work of the PACSCL team has discerned over the course of the grant, there are those collections which work well with minimal processing and there are those that do not.  Historically, we have never given the same level of attention to each of our collections.  Personal and family papers have often received more detailed processing than business papers and archival records.  While we have not adopted an MPLP approach at Haverford, we are interested in discerning ways of saving time and money while still providing rich access to our researchers and offering fulfilling and educational opportunities to our student employees and interns.  In the coming months we hope to try our hand at an “iterative” approach at enhancing collections by revisiting selected series within some of the collections processed to a minimal level under the PACSCL project.  And we aim to improve the remainder of our online finding aids bit by bit.

As one of the first institutions to dive into the PACSCL Hidden Collections project, we are pleased to see it wrapping up and hope that the other institutions who have participated have been as pleased and inspired as we have.