Appendix 4 • CLIR

Notes from Interviews with Archivists about Archon, Archivists’ Toolkit, Cuadra STAR/Archives, Eloquent, and CollectiveAccess

Archivists’ Toolkit Summary

To understand how archivists use Archivists’ Toolkit (AT), I conducted phone interviews with five archivists between May and July 2008. To encourage honesty, I promised anonymity to the interviewees. I tried to capture the interviewees’ remarks as accurately as possible, but I paraphrased and/or condensed some comments.

Reasons for Selecting AT

“The initial attraction is that we have a lot of tools in place for archival description and collection management, but they’re separate, distinct tools—data silos. We had the accessioning database separated from EAD database, along with a separate ILS, a separate database for A/V and photos, etc. The different databases were not integrated for end- users, just for the workflow point of view. People needed to learn various tools. It was difficult to reuse data because exporting demanded trying to cram it into whatever format the database was using.”
“We didn’t have a budget to purchase anything. We probably could have designed our own database, but we couldn’t have designed it to do everything that AT does. We could have customized things to meet past practice, but also decided to move away from old practices. We don’t want to be too flexible any more. Also, it was appealing that we could have input into development process as beta testers.”
“Our interest in AT is a function of where we’ve been with managing descriptive information and collection information—the information was all over the place. Some descriptive information resides in the card catalog, the library OPAC, and paper finding aids, and some in combinations. Accessions information until recently was done in paper form only, which made it difficult from a reference standpoint to locate that information quickly. We built a small database in InMagic around 1998. Location information is still managed in an Access database. All of that information was all over the place and still pretty much is. We had to look in all those places and had to keep those systems up. What I liked about AT was it was free, I knew some of the people involved in building it and trusted their judgment, and I felt like they built it with a lot of input from archival community, which has its pros and cons—it slows down development time, but hopefully it meets as many needs of community as possible. With the latest upgrade, they’ve added new stuff. Based on AT’s recent survey, they’re pushing at areas that we would like to see added to it. But we’re still struggling to fill in data for features they already have. Looking at it and seeing it demonstrated, it looked easy to use. And it is, particularly if you are familiar with archival terminology and descriptive fields. I liked the thought that we would be able to link our accessions information to our descriptive information. I liked that we could output easily to EAD. Our old system involved a lot of manual work. Now we can quickly spit out EAD or MARC. I haven’t done much with the print version of output yet but I think they are making improvements to that. That’s another feature that’s nice on the descriptive end.”
“This is the first thing I’ve seen since AIMS (?) in the mid-1990s that links accessions to collections and allows you to search accessions easily for stuff that’s unprocessed. One of the features in upcoming releases is the user tracking as well. Once we fully implement AT, we’ll be able to eliminate other resources, especially InMagic, which probably won’t be supported on future operating systems. I think it will reduce descriptive overhead for archives.”
“We had been looking for a management tool that would help us do some of the basic functions of an archive, such as managing our accessions, having a name authority and subject, and having some way of integrating finding aids into one tool. We’ve really been testing AT ever since learning about it. We’ve implemented parts of it fully, especially the accessions module. We are looking at or getting to point of implementing the authority module fully. We are still hesitating on the resource module, the place where we would import legacy EAD documents and create new EAD documents right in AT, export them, run them through our stylesheet—we’re still testing that. We’re hesitating because our legacy EAD documents are so diverse and weird. We have tested importing legacy docs and have seen what they come out like. AT is doing a lot better now with importing with 1.1, so we’re looking at going ahead and importing them. We need a stylesheet that works with exported AT finding aids and we haven’t quite that got yet. One of the things that we are considering is importing MARC records instead of the whole EAD, which would not only get around importing issues but also give us all the benefit of having our resource module linked up to accessions. There’s a way in AT to link accessions to resources backwards and forwards—there’s so much advantageous for us to have those resources in there that maybe a simple MARC record would be plenty for us to get subjects imported.”
“There weren’t a lot of archival management tools out there—we were looking more at database formats that were more or less homegrown. When I did research in 2005, I researched database structures in EAD and how things worked for people. I found a lot of different archives that had homegrown structures and found out about their limitations—we didn’t adopt any of those. We did hear about Archon and considered that along with AT, but at the time it didn’t seem to have as many possibilities as AT had for us. It didn’t at that time have a way of managing accessions—it was more a finding aid creation tool for small archives. And now it’s expanded a little. What concerned us a little about Archon is that it didn’t have ongoing grant support. We saw enough people adopting AT and felt that it had the solidity of ongoing grant support.”
“Previously we were using Access. There was no real way to get EAD out of Access, and we wanted to get finding aids on the Web. We were pretty pleased with what AT offered, especially EAD export. We have to abide by the Online Archive of California’s guidelines, so we needed to make some modifications to what AT exports to conform.”
“We’re using AT as a collections management systems—we’re not using the ability to produce finding aids. Within AT you have a resource record and component record [for multilevel descriptions]—a couple of different levels. We’re using it at the highest level to manage accessions and information about local collections.”

Ease of Use

“As someone who has taught an AT workshop twice, I can say that people pick it up pretty quickly. It does the basic things people need, and it’s easy to use for archivists who know what they need to do with archival description. Someone who wasn’t trained as an archivist had some problems with it; it’s set up with the assumption that you are an archivist.”
“We’ve been using the Resource module selectively. A few people have used it for finding aids because there have been special circumstances, such as needing to work off site, and it would have been difficult to set up our institutional macros and template. It worked out pretty well—we could help them get stuff online by exporting data from the Toolkit.”
“Archon and AT offer a good alternative to hand encoding. We couldn’t have trained [staff with a lack of technical expertise] in a reasonable timeframe to produce what they did with the Toolkit.”
“That’s a hard question. It’s not too difficult to use if you just need someone to input data into it. It’s pretty simple to get students to point to input data. But there needs to be someone in the department with a more thorough understanding of the program and how things work. Some things will need to be adjusted after the stuff is input; otherwise, you will run into databases that are not very standardized. The learning curve for all of the features of AT is pretty steep—it took me a month or two to get comfortable with it. Even now, I’m learning new things, such as digital object description or linking internally. I’ve trained staff and two interns how to input into it. They get information into AT, then I change things. Much stems from the hierarchical structure. It’s intuitive but confusing when setting things up. I have issues where they try to add a file to a box. In AT it’s not clear if a file is in box or equal to box. I have issues with structure and how AT displays it. As for training, I did a one-time 2-hour session for staff. Some picked it up quickly and jumped in; others took more time to get comfortable with inputting stuff into AT.”
“We have a lot of students working with data entry. It’s always a question of how much to give them. In my mind, the bigger question is how much organization of a collection can a student do. The students I’ve used are mostly undergrads doing data entry for legacy finding aids. They’ve been able to pick up on that. Most of them are fairly computer literate—the bigger issue is not boring them and making sure they pay attention to detail. What level of description you can train students to handle?”
“Seeing a tool like AT makes me wish I were starting an archives from scratch. Getting all of the old data into AT or any system is a challenge. We’re doing it piecemeal. Right now our main use is on the resources end—descriptive information, particularly for manuscripts at the collection level so that we can output to EAD and MARC. We’re fairly far behind with descriptive information, so that’s our big focus for this year. The plan is to get our accessions process in place at the beginning of next year. The trick is mapping our fields in our old database into the new database. With the new version of AT, they’ve got user-defined fields that will accept some of our oddball information—purchase price, appraisal value, in-house estimate of gift value, etc. But there are some data issues that are not straightforward that don’t map well, such as hard returns in descriptive fields, which cut the data off so that it doesn’t come across cleanly. There will be a fair amount of data cleanup to do to get it in there. Once the legacy data is in there, I don’t foresee any challenges to staff learning to use the system, either to in put or search data. I understand that getting accessions information is a challenge for everyone. It’s a little bit of a challenge with the descriptive information. It was stuff that was cataloged by many people over many eras using many different standards, or none at all. Cleaning it up will take time—but there are no significant challenges from the system itself. It seems to do everything we want it to do.”

Installation and Maintenance

“Installation depends on how you set it up. We have the back end—My SQL—set up on a server so various people can connect to it. Getting it set up in a networked environment took coordination from IT staff. Once that happened, it was smooth. We installed it on laptops during testing, and that’s been fairly quick.”
“Installation was pretty easy. We have a small systems department. One of our systems persons installed it. We just upgraded it, and that was like installing any piece of software. I don’t know what would happen now that we have user defined fields—what information would be lost with an upgrade. So far installing it has been a piece of cake.”
“Our systems department did the installation. On the listserv you see people with issues with My SQL. Our systems staff didn’t have any problems with the installation. We haven’t had any problems with the database. IT staff have moved it around a lot [onto different hardware], and it’s been pretty seamless.”

Ease of Customization

“There are built-in customization features. You can change labels of different fields, provide instructions or guidelines, etc. We have added look-up lists to add specific data and options.”
“In order to customize local use, you don’t need a programmer, just a set of guidelines to say, ‘On this screen, fill out these fields.’ For CLIR, it will be important for each repository to do the intellectual work up front of giving grad students good guidelines about how to formulate data. A lot of data is not in controlled vocabularies; there is a lot more loosey-goosey notes stuff. You don’t want to leave grad students up to own devices to put what they want where.”

User Community

“There’s a great AT users’ group listserv that is quite active where people ask and answer questions. We report bugs through the bug reporting system. We’ve found the developers to be extremely responsive to our concerns ever since the beta testing period. We’re very pleased with that; there’s a really good network of users built up.”
“The big thing about AT that will be interesting is that it will be leap of faith for institutions because it isn’t clear what the sustainability trajectory will be for it. We’re hoping and betting that it’s not just going to go away because we’re moving a lot of data into it.”
“My experience with the user support has been excellent. The listserv seems very active, and people don’t seem afraid to ask questions. You get a variety of people from AT responding to it. They seem to respond quickly, and they all seem to be on the same page. There’s not a lot of confusing dialogues. They seem to be able to handle both complex technical questions and simple questions. The manual that they created works well for me. The bug list that they put out is both helpful and confusing. They have a quickie style of documenting all of these problems. If you spend a few minutes, you can see that a problem has come up before. That sort of transparency about what the bugs are and how they are addressing them is a helpful feature. They have been active about doing presentations both at national and regional level. Without a huge budget, they’ve managed to do a lot of communication with interested users.”

Weaknesses

Potential problems with upgrading to new version of AT after making customizations
AT may be challenging for less technical staff to use. As one archivist commented, “AT is great project. I evaluated it and didn’t think that it would be as easy for archivists I know with limited technical skills to get it running and use it. It was a little too technical and required too much IT expertise to get the most mileage out of it.”
Lacks a public Web interface that would enable the public to search collections.
May not work with existing workflows: “We do use the resource module for some stuff here, but our general workflow predates the Toolkit.”
“There are still some bugs. It’s still not perfect, so some data may not be saved properly.”
“There’s nothing about it that has driven me crazy. The stuff that drives me crazy is that we have so much catching up to do and so few staff. AT is a significant improvement as a tool that helps us to get stuff done. I would like to see it link to user information. User tracking in AT would be good for part of our collections, but it’s not a holistic solution to knowing what people are using and where we should put our resources. But we have so much catching up to do that we’re not ready to implement that any time soon anyway.”
“In terms of resource description, I like it a good amount. The complaint I hear from my staff I disagree with. People say that it’s too clunky, it has too many fields, and you have to separate data into fields—to me, that’s good. People have gotten used to working in a Word document, without structured data. AT imposes restrictions, so it’s more of a mind-set of getting people used to thinking in a different way about what they’re doing in describing archival material.”
“Some more collection management tools would be nice, like doing stuff with processing priorities, ranking research value, current status of processing, level of description. There’s currently no way to track that within AT.”
“It’s hard for multiple people to work on describing one resource at the same time. They’re working on that in the next release: to merge different resource descriptions. If you are working on a huge collection with several boxes, it would be good to have people working on same collection at same time.”

“The exporting of EAD for AT is good; the exporting of MARC is pretty good, but not quite as granular as needs to be. It would be nice to have something that mapped to Encoded Archival Context for name records.”
“There were a few minor buggy issues we had with the first version, particularly with dragging things around, but those seem to be gone now. There are a few issues with this version where it seems to time out and lose data. Someone was working on collection, had the resource window open for half an hour, and lost the data. The Save functionality could be better so that you could save and still remain in the window. Now we save a lot.”
“I’d like the ability to rank collections, track processing priorities, states of collections, preservation, level of arrangement and description.”

“In general, I think the connection between the accessioning and resource modules could be a little stronger.”
“The problem with the import of legacy EAD is probably our biggest hurdle.”
“There are lots of places to put information in and you want to fill in every blank. You have to stop yourself from doing that and make sure that you’re entering what you need to and what’s necessary to create complete, valid documents that are DACS compatible.”
“The big challenge with AT is that it leaves a lot of options open to the user. You have to make choices, and there are lots of different notes available to you. What a grad student would need is for someone to say, ‘This is what we want to do’—that is, there should be guidelines locally to say how you work with his. You wouldn’t want to build the constraints into the software.”

Strengths

“The accessioning module is better than anything out there or that we could develop on own. We implemented the accessioning module first, and it’s pretty much what we’re using now.”
“The promise of having a single database for collection management. You do the accession record, push a button, convert to a resource record, and export as EAD and MARC. It’s not quite there yet, but it’s moving in that direction.”
“I actually like the fact that it is a database where people are forced to separate different data elements—it helps standardize data and produce finding aids quickly.”
“AT makes it quicker to produce finding aids.”
“I haven’t found anything better, particularly for the price. It’s a noble effort by members of our profession to fill a gap. It seems that they’ve gone about it the right way. Of the free products out there, they’ve got a good shot at keeping it going, particularly with the amount of implementations of it out there.”
“For collection management, I like the ability to produce reports about size of collections, different types, etc.”
“We’re very pleased with the accessions module and have been using some of the user-defined fields for our special needs. For example, we have needed a place where we could record material types in each accession, broad material types, whether visual, papers, digital items, etc. We have used one of the user-defined fields to enter that information. That will provide for us a way to use AT more as a processing planning tool. There’s a way to note whether each accession is high, medium, or low priority; we’re entering that information, so we can go through and find all the high-priority processing accessions in our collections and plan our processing from that. We’re hopeful that once we get that information entered into AT we can more fully use it as a processing/planning tool. Marking various material types will help people who are in charge of different media types—paper, digital, visual, etc. We can find all of those collections that belong to us and that way we can use the accessions records. Our legacy accessions database didn’t have a way to transfer locations directly into AT locations area, so we have to manually go in and enter all of the locations that we have noted, but once we have done that we can use it as a locations guide, so it’s going to be, and is already, an excellent tool for us. When you consider that we had very rudimentary accessions database in 2004, we’ve come a long way.”
“AT would be helpful for processing hidden collections. Right now we are cleaning up our accessions database. As we’re putting locations into locations guides, I’m finding some high-priority, ‘hidden’ collections.”
“AT has a business plan; there is a plan for ongoing operations that encourages us. And for us, personally, we have IT support that is really good. Should AT not become sustainable in future, we have ways in which we can seek IT support to sustain it on an ongoing basis ourselves. Sustainability is not as much a concern for us as it might be for smaller archives. But AT is looking at that and managing it pretty well for an open source tool.”
“We’re finally getting a place to put name and subject authority files and are really glad that we’re finally getting a complete accessions database. All that information is linked—names are linked up with accessions and resources. It’s a great tool.”
“It’s going to be a great way to plan processing. It’s one thing that has made our archives move forward with all of our management for our archives.”
“I think AT works really well. We had been thinking about using it to play around with producing METS digital objects. We’re in the process of doing mass digitization of archival collections—digitizing stuff at folder level and linking METS objects to finding aids and are figuring out how to create METS objects. The Toolkit is one of the things on the table. They’re supposed to be working on new functionality. Now you have to build whole resource description from the collection to folder level before you can build a digital object, but you will be able to build a METS object that isn’t connected to anything at folder level.”
“Not many tools are easy to use by people not trained in XML. If libraries have to train everyone who is working with collections to use XML, it will be challenging to roll out. In an XML editor, you don’t get a nice tree view; you have to do special things to produce that view. They are building AT so that you have metadata and visual screen that shows you where you are in the structure. In workshops, most archivists felt confident at end of 2 days in their ability to implement the tool. We need tools that work more like word processors and visually let you see where information is.”
“Ease of creating our resource descriptions. EAD export has worked fairly well for us. It seems pretty intuitive to use. It’s cut down a lot of work for us in getting things into EAD or MARC.”
“Down the road, I’m looking forward to having accessions and collection information interacting more.”

Archon Summary

To understand how archivists use Archon, I conducted phone interviews with five archivists between May and June of 2008. To encourage honesty, I promised anonymity to the interviewees. I tried to capture the interviewees’ remarks as accurately as possible, but I paraphrased and/or condensed some comments.

Reasons for Selecting Archon

The greater efficiency of using Archon as an EAD authoring platform: people creating finding aids no longer have to do it by hand and learn the EAD elements.
It is not as complicated as other systems.
Data is in standard formats that can be migrated into other systems should the need arise.
“We needed some sort of database that we could deliver to researchers for searching our collections. We wanted to have something that could go on the Web. Archon is free and pretty easy to implement without much IT intervention. … It gave us a quick and easy way to put collections on online, let patrons search them, and see everything we had, instead of having to search finding aids individually.”
“We needed something easy to implement for Web delivery of finding aids.”
“It’s open source. People at our library are passionate about open source. They are unhappy about contracts for our OPAC. Open source leaves us options if the user community is not active to continue the development ourselves.”
“The interface is easy to use, which is important since students would be doing a lot of the data entry.”
“There is a built-in Web interface that is an attractive, easy to use, out-of- the-box solution. We didn’t have to fight over what the system would be. Our systems people could do it, but this is out-of-the-box and we can just slap on our logo. All four archivists agreed on this.”
“I have more confidence in the sustainability of Archon. The University of Illinois developed Archon, they’re using it, they’ll keep supporting it for the long term, even if they didn’t have external funding.”
“This was an ideal tool for us because we had so little that was automated. We wanted to get information into the system quickly, using student labor. Students were just inputting stuff into intuitive fields. They didn’t have to know EAD and DACS.”
“An archivist here focused on EAD selected Archon. She felt that EAD is such complex work that she had to do everything herself. If we used Archon, which is simpler for someone without a lot of training to get started in, it would free her from having to tag everything herself. The result has been mixed. I don’t know if it’s really saved us time or not.”

Ease of Use

“Data entry is quite simple to learn.”
“Archon is pretty teachable. No software is intuitive, but the training doesn’t take too long. “
“Some institutions provide students and paraprofessionals with a cheat sheet that shows them what data to input where.”
“Archon is easy for nonarchivists to use—we quickly train students to use it. Like any other data entry, it can be tedious. With finding aids, the main difficulty is keeping track of where you are in the finding aid.”
“Archon would be a good choice if you’re a small institution without any Web finding aids, and you have students and volunteers. The great thing about Archon is that anyone can do it with 30 minutes of training. Scanning techs can cut and paste from Word into Archon. I might need to make sure that the intellectual structure is right, but it basically is easy to produce and go straight to Web. You can make changes really easy—with EAD, you have to change the file and re-upload.”

Installation and Maintenance

Installation of Archon is straightforward, but you may have difficulty upgrading it if you have customized your local installation.
“We did an upgrade 3 weeks ago and it was done in 15 minutes. There were a couple of hiccups, but it was smooth overall. Earlier problems were to be expected with a 1.0 release, and we could work around them.”

Ease of Customization

You can enter local information easily.
If you want to change the layout of your Archon site (e.g.. move around the standard elements on the Web page), you need to work with a programmer or Web designer.
With the current version, you can do a lot of customization through a CSS stylesheet.
One institution reported that the staff programmer didn’t like the installer code and decided to do the upgrade manually, since they had customized Archon.
“Graphically we’ve customized it—all the same information is there, but the fonts look different. Our customizations worked with the upgrade.”
Examples of customizations: Purdue, William and Mary

Weaknesses

Archon may be best suited for institutions that don’t have significant legacy EAD finding aids to import. Several archivists reported that they had trouble importing complex finding aids into Archon or that they were aware of this being a problem for other institutions. They were grateful that Archon staff attempted to help solve these problems, but ultimately one archive will have to manually redo several big finding aids.
You can’t enter formatting (such as italics) into Archon. “Archon is not Microsoft Word yet, since there is no easy way to format.”
Archon could provide better support for inputting special characters/ Unicode.
Archon doesn’t yet support structuring bibliography lists.
“It would be nice if you could take a box list and drag and drop it into Archon.”
“If all your finding aids are set up in the same way, they can be easily imported into Archon, but I know of no archive with that data. “
“It’s not yet possible to have different collections in Archon that have different ‘brands’—e.g., unique look and feel, search functions.”
“There are potential usability problems with the default layout of finding aids. Users may not know that they need to expand collapsed fields.”
“You can’t control the presentation of data as much as you can with a custom Web site. If you want to do anything fancy with the interface, you would need a programmer.”
“It would be nice if you could plug standard authority lists into Archon, or run a search of the LC authorities page and feed the results into Archon.”
“Although Archon recently went open source, it is currently being developed by a single institution. If they abandon Archon, then the user community will suffer. However, the developers use Archon and have a vested interest in seeing it succeed.”
“Archon is easy to customize, since it is based on CSS, PHP, and MySQL. However, it can be difficult to understand where each page is generated and what changes when you make a change.”
“Import/export tools in Archon seem to work well if you are moving data from one instance to another, but not piecemeal, one collection at a time. “
“Archon doesn’t support outputting content of collections in format optimized for printing.”
“I don’t think there’s enough guidance for users yet. I’d like to see expanded manuals for people trying to improve workflow and exporting to EAD/MARC. Right now, each place is separately trying to figure out how to change what it’s doing to fit into new system.”
Archon can improve its reports features, such as “report of accessions in last month, collections in one storage area. I hope that Archon will build that soon.”
“Our technical guy has said the PHP code isn’t very clean, but he’s not a PHP guy. Another tech person seems to be dealing with it fine. The Archon folks are working on cleaning up code.”
“I want to see some features become more robust. The accessions module is not as complex as would be helpful for university archives people. You need to be able to deal with annual deposits, accession number, date, etc. Archivists’ Toolkit does much better from most accounts with accessions.”
“There are little features that we want, such as the ability to hide parts of a finding aid for restricted materials. Right now you can have material either online or off, but it would be nice to hide part of a finding aid.”
“We’ve not yet used the digital library manager—we’ve heard that it needs to be more robust. I know someone who is using it and is happy with it. We have issues with loading our existing database.”
“Our main problem is importing existing EAD records. Archon is less forgiving than EAD; it’s like a database. If you tag EAD and it validates, you’re good to go, but Archon just won’t accept some stuff, such as IDs with characters (rather than a box/folder structure.) We haven’t been able to import three of our most important complex finding aids. We’re going to have to cut and paste these finding aids in. When you’re copying and pasting, there’s room for error. If it was just a mechanical import, I’d be more confident. We weren’t anticipating how to do that kind of work. The Archon people tried to help us but weren’t able to.”
“The Archon user interface is OK. The frustration with it is that you can only enter things a line at a time. If you’ve got a long finding aid, or if you’ve got something where things repeat, you have to cut and paste line by line, which is tedious. There must be some way to import it in larger chunks. My staff say they find it frustrating that you can’t see whole finding aid from the back end—in EAD, you can scroll up and down through whole finding aid, but with Archon you have to drill down through series, subseries, box, etc.—that’s all you can see.”

User Community

One archivist characterized the user support as “really good.” She typically e-mails the developers whenever she has a question, and they respond with enthusiasm. Archon has had a succession of strong grad students who have provided user support. Other archivists echoed the statement that the Archon developers are eager to help.
Archon does have a listserv, but it often centered on people who are just adopting Archon and lack technical support at their home institutions.
Fairly active listserv, with a few questions each week.
“The developers are incredibly helpful. There was an instance early on when I posted a question to Archon listserv. … In later release they added [the requested feature.] They have very responsive developers.”
“When I have a question, I have a really good response to it. I know some of the people involved; I have extra-strong ties with those folks. I e-mail a friend who is using it with questions. In talking to other folks who are looking at it, people have their eyes open about it, the good, the bad, and the continual development. I’ve found other users responsive to needs.”
“It’s good. It’s basically three guys at UIUC. They respond quickly to e-mails, but it’s not like a big commercial project; they want input for ways to improve Archon. I had a phone meeting with them on some problems.”

Strengths

Responsiveness of developers. An early adopter commented that everything she had complained about was fixed in later releases of the software.
Flexibility in working with different kinds of data. Archon takes any kind of media—sound, images, even a link to something else.
Makes capturing archival data more efficient. With the new accessions module, archivists can enter data into the system once and use it to generate multiple outputs. One archivist who hasn’t used the accessions module yet is excited that it will enable the archive to import standard data from an Access database and manage that data more efficiently.
Through the digital library module, archives can provide access to digitized versions of the objects described in finding aids, which researchers have really liked.
Web-publishing capabilities: All the data entered into Archon is immediately available online. As you enter data, it’s accessible to people live unless you ask that it not be made publicly viewable. At one archive, students enter data, but only the director of special collection can make it publicly viewable, giving things a final check and clicking a button to publish them.
Tools such as Archon and Archivist Toolkit may lower the bar for participating in EAD by enabling people to enter data into forms rather than having to know EAD coding. Archivists are embracing EAD over MARC because of the richness of the data.
Can create a draft MARC record that catalogers can then polish
Good authority control. As Archon has matured, the ability to deal with importing authority data and controlled vocabulary is coming along very well, which pleases librarians. Archon may offer EAC support once the standard is fully developed.
Makes information more widely available. For instance, Google indexes Archon contents. One archive reported increased interest in its collections from people around the world after it implemented Archon. Archon can make hidden collections more visible.
Easy to navigate. Everything is accessible in one stream; Archon feels like a Web page. When you’re in the Admin module, there are mouse-over menus in the interface.
Simple interface. Uses simple, easy-to-understand language. Archon is not archivist-centric, even though it is very useful for standardizing archival finding aids. It guides lay users through archival arrangement
“Users seem to like Archon—but we haven’t done user testing. After showing folks Archon in the reading room, we haven’t gotten negative feedback.”
“We’re pleased with its flexibility and power. We like how you can search at the top level, highlight results, and search within finding aids.”
“The ability to export to MARC and EAD is exciting for us. We have minimal cataloging support for MARC. To hit a button and have the majority of the work done is exciting. Especially for EAD—we don’t have the staff to do markup of finding aids.”
“I like how you can customize Archon—it’s easy to change the look of it. There are a lot of things you can do if you have some programming support. We’ve been using students to support customizations.”
“When I show it to people, we always talk about the out-of-box Web presence—it’s a really big deal to small institutions.”
“The browsability of it is great.”
“Everyone in the department can use it. With EAD, people who were using it had to go to 2 days of training. For people who aren’t working with it every day, it’s hard to remember how everything works. Archon is a lot simpler; I’m going to train our photo tech on Archon, then go in and set up series and subseries. I expect it will take 30 minutes to show him what to do.”
“Archon publishes directly to Web. You don’t have to deal with the systems department or replace each file when there needs to be a change.”
“Archon just added an archival management feature in its latest version. We haven’t used it yet (accessioning, etc.), since people weren’t sure if they wanted our collection management records to be all Web based, but I think it’s something we should try.”
“Archivists tend to like it.”

Overall Assessment

“Archon is excellent for ‘from-now-on’ or fresh creation of finding aids, but it’s a hassle to use with already-created finding aids.”
“Archon is the closest to a tool that allows you to only enter data once and have it come out in different forms you need.”
“Archon is new and evolving. They’re taking feedback.”

Archon’s Response to User Feedback

In response to user comments, Archon Project Manager Chris Prom indicated that some seem to be geared toward earlier versions of the software. Regarding the difficulty importing EAD files, Prom explained, “Since Archon has more restrictive data requirements than those of EAD, it will be impossible to write a single script to import every EAD instance. However, the current import script for EAD (PHP) could be customized by IT staff to handle difficult cases.” Version 2.2 provides better support for formatting data, and Archon has supported Unicode since version 2.0. For those who want to create separate skins or themes for different collections, Prom says that “the capability to do this is in the API, but it has not yet been implemented in the administrative interface. … A script to import authority lists from an Excel file” is planned, although “the link to the LC is more complicated.” Responding to the notion that the Archon code is a little messy, Prom notes that “we cleaned it up considerably, and have heard comments that version 2.0 is very well structured.” Prom also advises that improvements to the accession manager and digital library public interface are coming with Version 2.2.

Cuadra STAR/Archives Summary

To understand how archivists use Cuadra STAR Archives, I conducted phone interviews with three archivists between May and June 2008. To encourage honesty, I promised anonymity to the interviewees. I tried to capture the interviewees’ remarks as accurately as possible, but I paraphrased and/or condensed some comments.

Reasons for Selecting Cuadra

“When I was at SAA, I saw Archivists’ Toolkit and Archon presentations. I got excited about them—I’m a one-man shop with one assistant, a paraprofessional. I started to look into it [Archon?], but when we tried to install it, our IT group refused. They don’t support it and wouldn’t let us put anything on our computers that they couldn’t support. Then we went to the librarian here in charge of computers; when he saw that it ran on MySQL, he said no. So I started looking at other options. I’m not very tech savvy. We looked at ContentDM, but it was really for digital collections, less about managing administrative tasks and putting up finding aids. I liked Archon because it allows you to enter once and generate multiple reports. When we met with Cuadra STAR, we saw a demo; the electronic- resources librarian understood everything, and I understood everything on archives side. We both liked the service and liked it from the user side of things. It’s pricier than freeware, but they worked with us to find the appropriate price based on how many users can use it at one time. Since we’re small, we didn’t need many licenses.”

“We have lots of different types of materials—book, archival collections, history, A/V, etc. Our regular library system didn’t handle photos or archives well, but Cuadra has different modules to address these areas. Right now we are using MARC for library cataloging. We also have Star Archive & Star Images. We are just starting to use Star Archive for finding aids; we have a few in an earlier version of Cuadra software called Finding Aids. We’re also using Star Archives for a digital journals project.”
“I was not in on choosing it—but I think it was a choice based on flexibility. There were the most options available with Cuadra. The customer service was very good. They were helpful.”

Installation and Maintenance

“We run it on our own server. It has worked very well, and we have successfully gone through upgrades. The Cuadra folks put out nice instructions for updating. If there is a problem, they can help us right away.”

Ease of Customization

“Depending on what you want to do, the system is customizable. Originally it’s a database system, but they have made specialized modules to address different types of customers, such as information management for business users. You can customize it, modify data entry screens, Web searches, etc.—but there is a steep learning curve if you want to do that in house. I can do a lot of conversions and modifications in house. If you want certain changes, you can always have Cuadra do it.”
“We hope to host our own server in three years so that we have complete control of the Web interface. Right now, we contract with Cuadra Star to make changes to Web interface. If we installed it, we could use our own staff to make changes—it would allow us the freedom to have changes made in house. We’ve focused more on content and haven’t really put much effort into customizing the Web interface. It took a little while to refine how information would be stored or would display. We weren’t sure at the beginning what to ask for because we weren’t sure what the data would look like. We have started to do some customization. It hasn’t been difficult, but it’s a little slow—we submit a proposal, get a quote, get it paid for, and then it is changed.”

User Community/Support

“They provide excellent support—it’s very timely. When I had a question, the tech called me up, did a WebEx, and showed me what I needed.”
“There are help boxes next to different fields. One thing is sort of lacking—I’m a book person, and I would like a book, a user manual, a quick down- and-dirty how to. There is a book, but I didn’t find it helpful.”
“Cuadra is not a big company— it’s not like you call an 800 number. We can call to talk directly to someone who is familiar with our needs.”
“They have been responsive to problems. There have been very few problems with the server, and never for more than 24 hours in the almost three years I’ve been working with them.”
“Overall, it’s good. I think that the manuals and guides that they distribute are not very good. But they are very quick to respond and are happy to sit with you and help you through something. However, I wish they had more customer service reps with detailed technical knowledge. It would be nice to have closer access to tech support rather than sales. We funnel through one person. I’m the one person who speaks with the one person at Cuadra; they don’t want every person at an institution calling them.”

Weaknesses

“Cuadra/STAR is very specific, and I don’t have my finding aids in stand-alone files. I was just talking with OCLC about ArchivesGrid, but to participate I would have to export each finding aid as an XML file. I can’t get the export function to work. Some of the functions in Cuadra/STAR don’t really work yet. Sometimes it is limiting to be so contained within that one data management software. I can print out a finding aid, but I can’t do much else. If I have a patron and want to send them a container list, I can’t make an independent file. Ideally, I could select a collection, export the finding aid into an autonomous file of some type such as EAD or HTML, and distribute it independently of software.”
“There are a couple of pages where when you explode hierarchy out, you have to touch every single folder.”
“The person who set up my archives didn’t always do the hierarchy right, so I have to figure out how to put stuff into the hierarchy.”
“I had to get used to how it looks.”
“I don’t like the Web display very much. I wish that I could play around with it some. I can’t do that troubleshooting because we don’t have that freedom with them hosting our installation.”

Strengths

“They provide support—they take care of issues and host everything on their server. The data is backed up at their location, not here, which is good in terms of hurricane preparedness.”
“Since we got it, we’ll spend a few weeks intensely working with it, then spend time away from it. It’s not hard to come back to.”
“I love how you can search keywords in a Google way”
“You can cut and paste an entire legacy finding aid.”
“If I have the time, I could take a small collection, scan everything, and put it on the finding aid. I could look at everything associated with the finding aid.”
“Cuadra/STAR was designed I think by archivists. The terminology and hierarchy are familiar—scope/content, biographical/historical note—all the elements that you would put into finding aid template is there for EAD.”
“They’ll take your legacy finding aids and put them in for you.”
“I have only generated one EAD file. It was an easy, one-step process, but I don’t know what to do with it once I have it.”
“I use the accessioning function, which is fabulous and has so many different things to use. We don’t use all of the features. Let’s say you have a collection that people like to cite from and you need to give permission—you can add in every time something is cited and where. You could log donations, provide contact info for the donor and his daughter, and then switch her to the main contact if he dies. The sky is the limit; there so many pieces we haven’t used. If you move the record from accession to processed, you can move things over easily. You can even wait to make it live and flag it for release. They’ve thought of a lot.”
“You can maintain and enhance levels of description, from collection to item. You can set up and maintain repository data, do inventory control, manage circulation and loans, and maintain name authorities and the administrative interface. You can set accessions, review, and delete records. Inside the collection-level descriptions, there are all kinds of stuff. Once you figure it out, it is pretty easy. You can manage finding aids; you can flag records as ready to release and generate EAD. You can also generate MARC, I think. When you go into accessions, you can flag all kinds of things. It’s easy to search and pull up a record. You have an accession component, transfer settings, acquisition methods, value attached, donor, etc. You can put in all of the contact info, a credit line, and a brief description at accession level—everything that you might need. Once you actually do the top-level collection record, you can input basic information, such as collection level, display dates, arrangement and description, extent for finding aid, scope/content, top-level finding aid information, location, bibliographic summary, creator, etc. You can put in the authority level, history, retention, whether you expect accruals, date range, assigned location. You can assign it to a shelf. There is additional descriptive data—media, required technology (?), subjects, condition, acquisition source and ownership data, rights permission, access, reproduction rights, all kinds of stuff—as much or as little as you want to put in. You can get something up quick and could actually put in item level records in later. You can put in photos.”
“You can search by keyword or browse collections. When you pull up the result, you get a hierarchy on left with series level, and on right you see EAD.”
“I like the support. You know who you’re talking to.”
“It’s customizable to meet your needs—a system out of the box probably doesn’t meet needs that well.”
“I think it really gives me the framework for description—all I have to have is the data. I don’t have to worry about formatting or identifying my data because the software gives you so many options to fill in. With Cuadra Star, it’s very flexible, you have a lot of options, and you can customize how you present your information because there are so many options. “
“Cuadra is flexible in accommodating different types of media. That was its main selling point. We deal with many types of materials and it allowed us the freedom to describe to those materials. We have a staff interface, and a Web session for the public. There’s a link on our Web site to our public catalog—we can mount files of any type to records, so we can upload pdfs of docs, jpgs, maps, and we can also do MP3, wav for oral history, etc. Cuadra Star is appealing because you can upload files straight to the catalog and you don’t have to have a finding aid—it’s all linked together.”

Overall Assessment

“I’m happy that we ended up with Cuadra because I can get support when I need it—they e-mail you back almost immediately. There are no stupid questions to them; they provide very good support. The search interface is almost like a Google search. Especially as a small shop without much support, Cuadra is a good choice.”

“In general, I really like the system—it works well and is reliable and easy for day-to-day.”

Eloquent Archive Summary

To understand how archivists use Eloquent Archive, I conducted phone interviews with 4 archivists between May and July of 2008. To encourage complete honesty, I promised anonymity to the interviewees. I tried to capture the interviewees’ remarks as accurately as possible, but I paraphrased and/or condensed some comments.

Reasons for Selecting Eloquent

“We selected Eloquent back in 2000 when they were using GenCAT, a DOS based system. It’s reasonably tailorable. We can make it work for our particular needs rather than changing our practices to fit a system. It has worked well. At beginning, none of the information was in an electronic format, so we had to do a lot of data entry. It moved from DOS system to Web-based system a couple of years ago—they did all of the migration for us.”
“I used Eloquent many years ago on another project and liked it at that time. When I got the chance to buy software 10 or so years later, I looked at other companies and once again Eloquent was the one I chose. It was the one most likely to do the job. One selling point: the data conversion from our old system (Filemaker) to the new one was less expensive with Eloquent than with other vendors. They delivered the converted data on time and with good results. We also chose Eloquent because it is Web based. I can log on anywhere at our facility.”
“We already had an Eloquent system in place, so they were very familiar with our data and data structure. That familiarity facilitated a great deal of things.”
“GenCat was chosen as the archival descriptive database back in 1996. We used GenCat until 3 years ago when it was experiencing difficulties, such as corrupted data. We weren’t happy with some of the support we were getting from Eloquent. We did a review of different software available at that time; we looked at MINISIS. We also looked at ContentDM as means of holding of descriptive data, but our IT people said it might be useful for description, but not for other purposes because it wasn’t relational. Because we weren’t totally happy with the other options, we liked WebGenCat better than anything else at the time. We’re happy with some bits, but not so happy with others. We’re using 3 modules: library, archives, and records management. The records manager has not been happy with that module and may look to something like Documentum. We’re happy with the library component that we’re using with theses; it’s flexible, shows the records extremely well, and is easy to tailor. We didn’t purchase the part that would allow us to tailor the archives module, so we depend on Eloquent to do customizations. We’ve found that the library component is superior to DSpace. For the archives module, it has real possibilities that we have not realized.”

Ease of Use

“It would be easy to use if you train grad students to do it—especially if you have Eloquent do the work for you.”
“When we first installed it, we had a clerical person who had a difficult time understanding the hierarchical structure and the language because she didn’t have an archives background. I think that paraprofessionals and grad students would be OK, but the system does presume that you know archives somewhat.”

Ease of Installation

“It was easy to install; it took less than an hour. As for maintenance, there has been basically none. The only problem we have is that from time to time the system hangs and we have to restart it. We can’t figure out what causes it. It hasn’t been a big enough problem yet that we’ve invested time in solving it.”
“They’ve been very supportive throughout the entire process, from migration to installation—they worked with us very closely and slowed down to my speed. All in all, I don’t think it was that difficult. The timelines we initially set were probably not as realistic as they should have been. They were very willing to work with us. All in all, it was a smooth transition.”

Ease of Customization

“One other archivists and I are the administrators and so we can do the tailoring for ourselves. We create our screens so that they fit archival standards and what our users are used to.”
“We have done some customizing. We did purchase the architect component so that we can do some stuff here. When we use it, we usually have their support on the phone with us. They’ve been very helpful in terms of walking us through various changes. When we used Eloquent’s DOS system, we had tweaked our system so much that when any upgrades came, we couldn’t do the upgrades any more. This time around, we went with the system based on what they had, and they were able to implement all of the stuff we had changed. It went pretty smoothly.”

User Community/Support

“There are no training manuals, so we need to figure out what they’ve called things and how the scripts run. We’re learning how to do things.”
“User support is really responsive to questions. The main drawback is that they do not have a manual—often there isn’t anything in the help notes.”
“The service aspect is weak.”
“I had an assistant who did the most of the interactions with Eloquent, and in general we found it to be good. It sometimes takes a couple of hours or a day to get a response, but on the whole user support has been very good.”
“User support has been excellent. They respond almost immediately. They offer to walk you through something. They’ll provide detailed instructions via e-mail or the telephone. We’re looking at the same thing as the changes are happening. Their customer service up to this point has been really great.”

Weaknesses

“We have to do communication with the company by phone or pay someone to come here. It would be nice if we had an in-house system so we would have someone to fix problems here.”
“We have had big problems working with Eloquent in getting what we needed adjusted to suit our needs.”
“The behind-the-scenes things like creating your own report or importing and exporting can be somewhat difficult. We do have the Architect’s module, but that sort of work is still difficult to do.”
“We’re unhappy with basic reports. It seems that there is basic information that any archival institution would need when doing a search, such as an accession number. Some reports don’t have the basic information you need. For instance, the collection lacks a file number, which is basic information that any repository would need. When we ask them to make changes, it just doesn’t happen for a long time. … There’s been a lot of frustration.”
“We’ve been trying to get them to export metadata for one of our collections so that we could put it into ContentDM. In the old version of GenCat, you could import and export data easily. We’ve been trying to get this data out of Eloquent for about a year. We can’t do that in the Web version and having trouble getting a response from them.”
“There is no written documentation. Some of the architectural stuff is difficult to do. They really could use documentation.”
“There’s not really a weakness that comes to mind. When we typically have a problem, we either e-mail or telephone the help desk and they’re ready to help us. You always find quirks when you start something, and they expedite everything and tell you how to do it. Merv Richter, president of Eloquent, was involved through all of the steps, ensuring that his staff was there to help us. They pretty much held our hands through the entire process.”

Strengths

“We use it for all of our workflow—receiving, accession, creating descriptive records, tracking researchers, appraisal, authority control, retention schedules, etc.”
“You can create EAD with the system, but we haven’t done that yet. Supposedly you can push a button and automatically generate it.”
“Comparatively they are cheaper, at least when we were initially looking around.”
“It’s tailorable. We’re not stuck with an out of the box model—that’s our high point. We envision sticking with them for next several years, since the system fits everything we wanted to.”
“I think the product itself has a lot of potential. I liked GenCat, I like Web GenCat OK. There are hotlinks to subjects and authors.”
“What’s nice about the system is that it would be easy to export—every field is delineated and it’s straightforward where the data lives.”
“It’s very easy to use and does exactly what you would want in an archival system. An archival system is really quite difficult in its organization. We had tried to design our own, but to have all of those problems thought out in advance is very useful, since Eloquent includes features we didn’t even imagine we needed.”
“I like being able to make changes and immediately post them to the Internet. I like the immediacy of it. If someone discovers a typo, we can immediately make the change and post it. We can digitize something and immediately attach that file to its description. I know a lot of folks use Content DM, but Eloquent allows us to attach images immediately to descriptive record. Everything is in a single system. We can link any descriptive item to a digital image, PDF, mov file or whatever, and have it displayed on the Web immediately. Reference requests have more than doubled as a result, which is something we’re struggling with now because we’re short staffed. It’s definitely improved our Web presence.”
“Right now Eloquent is doing what we need it to do. They met me where I was at and they really slowed down the process so that I could get on board. They worked with us closely from the data mapping to migrating the data to implementing the software. Merv Richter himself came down to do the training. We walked through all of the screens and all of the configurations. That made for a well-rounded experience from beginning.”

Eloquent’s Response to User Feedback

In response to user comments, Eloquent’s president, Merv Richter, gave the following (slightly edited) response:

Some of the negative comments may have come from customers migrating from the old DOS version of the Eloquent GENCAT product. Applications built with it were usually custom-built, so when moving to the Web-based packaged application, some of the personalization was lost. Also, some chose not to purchase the WebGENCAT Toolkit for the new product, so they had to pay for custom tailoring to their new application. The package product would not accommodate the old data structure.

Customer service is available by calling the hotline. Those calls are usually resolved by the person answering the phone. Voice messages are responded to in less than two hours.

The Eloquent Archives application is delivered with utilities to export data in a number of formats including ASCII tab-delimited and Excel. Eloquent consultants can configure a custom export to gather all related data out of the database and string the data fields out in any sequence the customer requests. The customer then uses the powerful search tools to select the desired records and send them to the custom export utility. The entire project usually does not exceed three to eight hours of the consultant’s time after the customer approves the format.

CollectiveAccess Summary

To understand how archivists and museum specialists use CollectiveAccess (CA; formerly known as OpenCollection, or OC), I conducted a phone interviews with two users between May and June 2008. To encourage honesty, I promised anonymity to the interviewees. I tried to capture the interviewees’ remarks as accurately as possible, but I paraphrased and/or condensed some comments.

Reasons for Selecting CollectiveAccess

“We are using it for a couple of things. The features that were most appealing were the complexity of relationships that you could catalog between individuals. The project we’ve used it to execute is mainly a database that combines collection objects, artists and story, and locations. It has a built in georeference capability that is easy to use. We had planned to use more traditional collection management software for our project, but when that didn’t work we turned to OC.”
“A lot of the other software that was looked at was too expensive. OpenCollection is free, a big factor in why they chose it.”

Ease of Use

“It’s much easier than traditional collection management systems that I’ve worked with.”
“It would be easy for someone who isn’t a trained archivist to use. As long as there is a protocol written for someone who is entering the data, it’s pretty straightforward. The only thing that might take more explaining is the taxonomy that you create. The data is pretty easy.”

Ease of Customization

“If you have someone who can write code, you can do all sorts of things. Even I can change the names of fields, check boxes, etc. You can make it do what you need to do.”
“It’s very flexible and customizable. Any time you need to add a classification, it makes it really easy; it has an easy interface.”

Weaknesses

“There isn’t a lot of documentation—no help manuals.”
“The only weakness is that you don’t get a help desk. You can e-mail OpenCollection and they get back to you.”
“Sometimes there are certain things on the interface I find a little bit clunky, but I’ve given a lot of feedback, and changes are made quickly. Whenever I find something that is awkward, I’ll e-mail Support. Since it is so customizable, they can change it. I haven’t really used other archival software, but I know that it has been very easy.”

User Community/Support

“I’ve found that the support has been very helpful. It’s been really easy to access people.”

Strengths

“I’m not someone with a lot of experience with these systems, but I like the ability to link objects to people to places to events to exhibitions. You can use it to tell stories and show relationships between things. The end product lets people navigate through those relationships.”
“It’s visually very strong—there are a lot of visualization options.”
“It’s online and customizable—you don’t need to download anything. Any changes that I make can be see instantly by anyone else working on the project.”
“There are cool tools within the software. For digital photos, you can magnify photos at a high resolution so you can catalog it at a level that is really detailed. There are details in old photos that you can’t see through a magnifying glass, but the zoom tool on the photo interface is really powerful. The same goes for oral histories. … All of the audio files are digital, so you can catalog separate pieces of a whole oral history and jump to that part. The level of cataloging lets you get to minute detail. It’s really easy to use.”
“Authority control is pretty good. You can connect to any sort of authority. We’re connected to the Getty Art and Architecture Thesaurus, and it’s pretty good. If you can’t find something or if it doesn’t fit, you can create your own authority. In that way, it’s nice because it’s customizable. It’s flexible.”