Wednesday, January 28, 2009

Our Evergreen Success Story, Part I

In June 2008, we announced the migration of our library automation system to the open source ILS called Evergreen. The news appeared for our patrons in our online newsletter and for Maryland Librarians on page 7 of the Summer 2008 issue of The Crab, and as a press release by Equinox Software, but we neglected to mention here that the migration took place.

Now, almost 8 months post-migration, as I prepare to speak at the Computers in Libraries 2009 Conference in DC (Presentation E204 if you're going) I realize the omission. I blame maternity leave and a host of other circumstances... but that's neither here nor there. Without further ado....

Kent County Public library is now live on Evergreen and has been since the great (to us) migration of June 4, 2008. We set several target dates for the migration and performed some test migrations before we took the plunge, but we beat our deadline by nearly a month and had what I consider a smooth transition from Horizon to Evergreen. We haven't looked back.

Since we are a small public library (19,500 population served, 150,300 yearly circulation) with limited funds and limited in-house technology staffing, we elected to outsource the support and hosting of our Evergreen installation to Equinox Software the Evergreen experts. The migration was handled by Alpha-G Consulting.

Our reasons for choosing Evergreen in the first place have been outlined in detail in earlier posts, but in a nutshell, we liked the freedom, the features, and the price.

We liked the freedom and options that come with open source software. Specifically, we're not locked into a particular vendor for support, hosting, or enhancements. While we may never have to make use of this freedom, it's good to know it's there. And in case you're wondering, after 8 months live, we're still happy with Equinox Software, our chosen vendor. The support has been excellent, and the response time amazing.

We liked the features already existing in Evergreen, and those planned for future releases. When using Evergreen, it's evident that a lot of thought has been put into how the software can best accommodate the workflow of the librarian, instead of the librarian accommodating the workflow of the software. While we are a single library, and not a consortium, we liked the fact that Evergreen was designed to work well in a situation with multiple libraries sharing a database of items, but still retaining their independent governance and policies. Someday we might like to be involved in such a resource sharing consortium.

We also liked the price. Our chosen vendor worked with us to make Evergreen fit within our budget, and we wound up getting a lot of bang for our buck. Our situation aside, the very nature of the open source software market is more economically favorable to libraries than is the proprietary market. In the proprietary market, one vendor holds a monopoly over their own product, allowing them to set whatever price they choose, with fear of migration helping to retain customers. In stark contrast, in the open source market there is no licensing fee, and anyone can set up shop offering support, hosting, etc. for an open source software product, so the competition (or even the possibility of competition) among vendors for customers naturally acts to drive prices down to a fair market level.

Please stay tuned for more details on our migration process and our post-migration experiences.

Karen Collier
Public Services Librarian
Kent County Public Library
www.kentcountylibrary.org

Friday, March 7, 2008

Evergreen at PLA

Want to learn more about Evergreen and meet others involved in the Evergreen community? Attending the Public Library Association's conference in Minneapolis (March 25-29, 2008)? The Evergreen community is planning a "Birds of a Feather Gathering" that you won't want to miss. Andrea and I are planning on attending, and we hope to see others from Maryland there as well.

March 27, 5:30pm - 7:30pm at the Minneapolis Marriot City Center, Gray & Wayzata rooms (8th floor).

Wednesday, January 9, 2008

The Advantages of Sharing a Database

In an earlier post, I described how Evergreen's flexibility makes it ideally suited to sharing a database of patrons and materials among multiple library systems. The point of this post is to explain why libraries in Maryland might want to do that. I see three main advantages, which I will discuss in more detail.
  1. Seamless inter-library loan
  2. Statewide library cards minimizing data re-entry
  3. Cost savings
Interlibrary Loan - The current interlibrary loan system used in Maryland is not perfect. At best, it requires a patron who is looking for a book to know that there's a second place to search (Marina) and to go to Marina, re-enter the search, and place a request. At worst, for some counties, patrons cannot place their own requests, except through the intervention of a reference librarian. Just to add insult to injury, the Marina interface is slow and provides only the barebones bibliographic information about the items it lists.

In contrast, for libraries sharing a bibliographic and patron database, the interlibrary loan system would be seamless for patrons and staff alike. Patrons would need to search only once in the online catalog, and could choose whether to see only items in their preferred branch or system, or to expand their search to see items throughout the state. If the patron wanted an item not available in their own system, they would request it through the catalog, as they do now for items in their local system. Staff members would have less processing and data entry to do, since both the sending and receiving library already have access to each item's record, so books could be sent between library systems as easily as they are now between branches.

Statewide Library Cards - Maryland currently offers its residents a statewide library card called an MPower card. This card can be used in any library in Maryland, regardless of which county you live in or where you first got your card. What many people don't realize though is that you still have to go through the full registration process the first time you visit a new county library system, providing all your information, showing identification, etc.

However, if the county library systems in the state were to share a single database of patrons, registration in a new county could be as simple as showing your statewide card, so the library system knows to count you as a patron.

Cost Savings - A white paper by New Jersey academic libraries does a good job of explaining the cost savings in using this type of open source software over a proprietary system. Because Evergreen runs on the Linux operating system and PostgreSQL database, neither of which require expensive licensing fees, and because it can run on less expensive hardware than many proprietary systems, there are substantial savings to be had. Not to mention, if you're combining the data of many library systems, economies of scale help out considerably. There's less infrastructure required for a shared system than for many independent systems. Similarly, one would expect lower support costs if support is outsourced because it's a single large system rather than many smaller systems. Not only that, but there's the option of hiring local staff to support and improve the ILS.

Georgia public libraries first developed and put Evergreen to the test in a statewide environment with a shared patron and bibliographic database, with much success. Public libraries in Michigan, Indiana, and British Columbia and academic libraries in New Jersey are at various stages in exploring or implementing the possibilities Evergreen and a shared catalog and patron databases could have for their regions. Why not explore the possibilities in Maryland?

Tuesday, November 20, 2007

Comparing Evergreen and Koha

As mentioned in an earlier post, we at Kent County Public Library explored our Open Source ILS options, setting up servers running both Koha and Open-ILS (Evergreen), the two main open source options. We read their web sites, tested out their interfaces, poked around in some code, joined their mailing lists, asked questions, and got quotes for migration and support. Liblime is the main support option for Koha, and Equinox is the primary option for Evergreen.

On the whole, we were impressed by both products. I felt that both systems easily rival the products of the big name companies in integrated library systems in terms of functionality and usability. Both systems have large communities of users and developers actively working to improve the software, according to the needs of their users. And support is available for migration and ongoing usage, not only through a contract with one or more supporting companies, but also through the community of users. And no, this isn't just theoretical, I've already received a lot of help and answers to my questions through the community mailing lists.

Since both these products are continually evolving, any comparisons must be taken with a grain of salt. And since they're both open source, it's entirely possible, and even likely, that they will cross-pollinate and each will take on features of the other, within the constraints of their basic frameworks. That said, here's my attempt to compare the two....

There are two main differences that tend to matter most to people considering an open source ILS. Koha already has functional Acquisitions and Serials Modules, while for Evergreen these modules are still under development. Evergreen has more administrative flexibility, which among other things means that sharing a single catalog among multiple library systems in a consortium is practical, if you wish to do so.

If your library needs acquisitions or serials modules and needs them now, then Koha is probably the way to go. If you can get by without these modules, or you can stand to wait a little while for them, then I would recommend giving Evergreen a very serious look.

As I understand it, Evergreen's greater administrative flexibility makes it possible to set circulation and holds rules at the consortial, system, and branch levels, meaning that these decisions can vary from location to location, rather than having to force a consensus among independent decision makers. Similarly, permissions can be set for groups of staff and individual staff members, and among groups of patrons, giving everyone the ability to do exactly what they need to be able to do, while minimizing the chances of unintentionally changing something related to another branch or library system that they don't need access to. So that's how Evergreen makes a shared database within a consortium practical. Look for a future post on why a shared database might be advantageous.

It's tempting to overgeneralize and say that Koha is for individual libraries while Evergreen is for consortia, but that's not true. Evergreen has features (described above) that make it ideal for consortia, but for an individual library that won't necessarily make use of all its features, there's no harm in having them, and they could come in handy someday.

Another consideration is the look and functionality of the OPAC (online public access catalog), which patrons use to find books and other materials. Demos of the Koha Zoom (newer interface for Koha) and Evergreen interfaces are available online. As you explore these interfaces, try to keep in mind that the look (colors, images, layout, even text) is very easily customizable compared to functionality which can be trickier to change.

Personally, I love the feature in the advanced search of Evergreen where you can group formats and editions, and would like to see that as the default option for all searches. On the other hand, Koha has a Federated Search feature, which is a good idea, but could stand improvement in its implementation. In its current state, rather than offering federated searching on your search terms when there are few or no results, you only get the option to do a federated search for the title of an item you've already found in the catalog.... Hopefully this feature will be re-worked in Koha, and perhaps Evergreen will add a Federated Search option as well.

I have tried to fairly represent the comparison between the two software systems, but I'm sure my personal preference has shown through to some degree. If it sounds like I like Evergreen better of the two, it's because I do. Evergreen is the system that the Kent County Public Library has decided to use, and I'm very much looking forward to the switch.

Wednesday, October 10, 2007

Open Source ILS

We have been dissatisfied with our Horizon Integrated Library System, and have been exploring alternative options for quite some time, particularly since the demise of the promised Horizon 8. Most of the other ILS companies were either out of our budget or didn't seem to offer exactly what we were looking for, particularly in terms of customizability.

Which led us to explore Open Source as an alternative option. Support isn't free, but competition between vendors supporting the same product can drive the price down. And if there are features we want that aren't available yet, we can contract with someone specifically to develop them for us rather than rely on the magnanimity of a large company to listen to what we, a small library system, want. Not to mention the ideals of the Open Source community dovetail nicely with the ideals of the library world.

So, we explored our Open Source options, setting up servers running both Koha and Open-ILS (Evergreen), the two main open source options. We read their web sites, tested out their interfaces, poked around in some code, joined their mailing lists, and asked questions. Then we contacted vendors who do migration and support for quotes. Liblime is the main option for Koha, and Equinox is the primary option for Evergreen.

We were impressed with both products. Watch for future posts comparing what we found about these two options, the decisions we've come to, and why.

Friday, September 14, 2007

Google Mini

One of our early projects, which is still active, is the Google Mini. We purchased a Google Mini Search Appliance, which is designed to index Web pages and other types of files on a corporate server, web site, or intranet. These documents can then be searched through an interface like the Google search everyone knows and loves.

We had two main goals for this search appliance, to index the content on our Web site for easier access, and to experiment with providing a new interface for patrons searching our catalog. In terms of the catalog we were intrigued by the dual possibilities of "Did you mean...?" style spelling suggestions, and Google's usage based ranking algorithms applied to our library's holdings.

Jerry first found the idea in use by Danbury Public Library with their III automation system. We set out to see if we could accomplish the same thing with our Horizon ILS system. The result can be seen at search.kentcountylibrary.org.

I will provide more details on how this was accomplished in a future post if readers are interested. The degree of success and usefulness of the project is still being evaluated. We'll keep you updated. And meanwhile, feel free to let us know what you think.

The Cast of Characters

Our director, Jerry is a visionary and enthusiastic leader, who has confidence in his staff to accomplish whatever we set out to do.

Andrea is our Technical Services Librarian, Cataloger, and Systems Librarian. She has a great deal of interest in databases and information storage and retrieval theory.

Karen is (I am) our Public Services Librarian and Web Developer, who is ever eager to expand her techie skill-set. Recent projects have involved online interfaces, Linux, and PHP.

Bob, occasionally on loan from a neighboring library system, is a source of inspiration and information, with his extensive knowledge of SQL and Library ILS systems.

ESRL Techs keep the computer networks running for our library system and 7 others on the Eastern Shore of Maryland, and help us out in numerous other ways. Their support is indispensable.