Bess Sadler: Open Source ILS and Library-in-a-Box
Posted by odyssey2007 on April 20, 2007
Bess Sadler is Head of Technical and Metadata Services at the University of Virginia and co-chair of eIFL-FOSS.
What is eIFL-FOSS?
Electronic Information for Libraries—Free and Open Source Software
- eIFL is concerned with providing and developing access and support for electronic library services in developing and transitioning countries around the world
- providing alternatives to vendors and proprietary library systems
- contributing to the overarching trend in open access and open source: “Libraries doing it for themselves”
- library members who are part of eIFL-FOSS are often seriously under-funded and have inadequate services
Open Source Software and Library Systems
- what is open source?
- after writing a computer program, you have to compile it; vendor software is packaged and already compiled which means you can’t make changes
- free software is not always what it sounds like; there’s a difference between something that doesn’t cost any money (as in beer) and something that allows you to participate freely (as in speech); and something that is easy to come by (free: as in kittens) but has maintenance costs and long term commitment
- UVirginia and eIFL-FOSS are interested in avoiding licensing fees, but willing to pay for developers and maintenance
Building Open Source Tools
- Open Source software already exist and some products actually allow you to build tools e.g. Linux, Firefox, OpenOffice
- the best building tool is “community”; without considering community of users and other builders, you are only building tools for yourself
- harness the developer community, conferences like the Lucene Summit that bring together Open Source software tool developers
- the kinds of tools that eIFL wants to use and supply and support in developing countries are library software like Koha, DSpace, etc
Who is eIFL?
- an international consortium of libraries who are interested in developing and adopting open source systems software
- eIFL is funded by Open Society Institute (OSI) and Soros Foundation whose goal is to allow developing and countries in transition adequate access to information for education research, the economy, and institutional and personal development based on the belief that ultimately there is a chance for democracy when information is accessible
- OSI focuses on promoting free speech, freeing prisoners, etc; as well as improving access to resources that keep the people of those countries up to date
- eIFL is involved in teaching under-serviced member librarians how to gather better resources using a unified voice through:
- negotiation skills
- consortium building
- and knowledge sharing
- eIFL-FOSS is focused on advocating that communities who are self-sustaining rather than building reliance on external aid or short-lived economic support through:
- the use of Open Source and developing OS tools software
- building a community who are aware of how to use and develop OS tools
- the philosophy behind Open Access is about ethics: using public funding to provide public access to information rather than taking public funding and creating private, secretive information
Barriers for Libraries in Developing and Transition Countries
- in Zimbabwe, currency does not have longevity which means that relationships with commercial software vendors are difficult because the “inflation rate of above 2000% poses significant (and often insurmountable) challenges in purchasing foreign goods” (Sadler, personal communication**, 2007); however, a self-sustaining OS system overcomes this issue of access
- in Georgia, no ILS system will support the Georgian language because those systems are unable to create individuals modules that can be customized to the needs of the community that is not a major language
- most of these countries are recovering from colonialism and trying to reinstitute their own culture and knowledge sets, e.g. Georgia has just been released from Russian rule, but the primary language of most systems are all still in Russian
- modeled on the NGO-in-a-box software from Tactical Technology Collective
- Library-in-a-box aims to be an Open Source ILS that can be integrated, fully internationalized, easily distributed, and installed; internationally supported interface and documentation; self-sustaining systems
- existing options include Evergreen and Koha
- there’s no point in creating an ILS that recreates something from 10 years ago or reinvent a system that creates the same problems that librarians face right now!
- implementing the “Leapfrog Effect”: developing the software locally means the project also creates local education and job skills, so the trickle effect is that the local infrastructure is also developed and improved because people are trained and can communicate with other developers globally and new interrelationships are created
- for example: Malawi had a 5-year grant from a Japanese funding agent that supported vendor subscriptions, but now the funding is running out, the library is worse off than it was five years ago because they don’t have any sustainable systems e.g. a card catalogue
How do you build it?
- but how do you build community?
- it’s not easy
- TacticalTech.org is a partner of eIFL
- TacticalTech helps NGOs build their technology according to their specific needs—e.g. privacy protection and security—libraries are essentially NGOs in the big picture
- face-to-face meeting is an invaluable way to build community
- when starting a new project, TacticalTech enters a region, asks everyone to gather and then asks them to say what their needs are
- then, launch the development project using local tech, not external staff
- identify and harness the strengths of the local people and then delegate and encourage them to take on the technical parts of the project-everything from working on refurbished hardware to loading Open Source software
- eIFL is interested in creating partnerships between countries that have similar language usage—English, French, Russian and Arabic—grouping is determined by the language that will be the primary development language
eIFL-FOSS Current activities
- studying the current state of the OS ILS available using the Business Readiness Rating, including both versions of Koha and Evergreen and maybe others to compare and design the next OS ILS
- Evergreen, for example, is internationalizing, so everyone benefits from their initiative and funding
- the beauty of OS is that these kinds of initiatives, like translation, become shared—no one has to reinvent these features
Project BlackLight ILS
- Project Blacklight is an OS ILS being developed by a group at UVirginia, it resembles Endeca’s faceted searching and uses Lucene; the experiment version it holds over three million MARC records
- Blacklight retrieves materials quickly according to relevance because relevancy ranking is tuned by the user by choosing format options etc (DVD, book, etc), library collections (by library building), by genre (correspondence, fiction, drama, etc), by topic (e.g. country music)
- also, if unique keys match then you can cross-reference to other catalogues (i.e. the UVirginia SirsiDynix catalogue system)
- this tool was created with basically zero resources; dedicated staff time was the only thing that was unaccounted-for expenditure
- it took them a few evenings to build this; and they hope that something like it will be the basis for Library-in-a-box
How can Libraries Do It For Themselves?
- internal grassroots skill set training
- UVirginia offers a weekly, voluntary programming class
- e.g. people need to know XSL to work on the digital library system, and the participants learn XSL; Beth gets the participants to work on projects she’s doing, and they learn how to do it
- allow people time, staff, and the tools to “play”—ultimately this has to be advocated to the decision makers as a money-saving initiatives in the long run
Q: “Does eIFL work only in countries that are democracies?”
* eIFL “do not make distinctions of that kind when countries want to join our consortium” (Sadler, personal communication**, 2007)
Q: What is the Business Readiness Rating?
* “A Framework for Evaluating Open Source Software”: the BRR is one of many standardization model for assessing open source software so that when people want to take the initiative to build their own tools and software, this is a system that alleviates a lot of the trial and error time and effort usually taken by trying to decide what software to use
Q: I’m involved in Public Services and don’t do any coding: how do I get involved in something like eIFL?
*the best way to contribute is to contribute to the Open Source community, developing tools, advocate for Open Source solutions
*eIFL also need developers who are doing beta testing, attending vendor demonstrations and reporting back to their home organization and also advocating for Open Source alternatives
Q: What kind of advocacy is happening in the developed world as well?
*advocacy and outreach is a primary occupation of eIFL members in the developed world
*developing more open source and promotion of open access alternatives
*whole institutions are taking on the challenge of adopting Evergreen as their primary ILS; again, if Laurentian University adopts it, they will translate it to French, which means that the entire community benefits
Q: How would a library go about adopting or experimenting with an OS ILS?
*no, there are vendors in the OS realm as well
*for example: a company called Equinox Software designs support systems who want to transition to something like Evergreen
- so even if you don’t have the language or programming support, this “vendor” can help the institution make the transition
- the beauty of this kind of “vendor” is that because the software is OS, the relationship with the vendor is non-exclusive
- anyone can start a company to support an OS package
*or, get OS software packages and try them out—you are under no obligation to the vendor to do a set-time trial; you can get them up and running quickly and play with it for as long as you want
*or, attend Code4lib (2008 is in Oregon) and attend an on-site training workshop
Q: What does “internationalizing the interface” entail?
*traditionally, the interface is often fed by hard-coded strings of text, but the interface code should be created so that a translator, for example, can download the interface strings of text, create those same strings but in translation, then it gets plugged back in
*remember that the interface design must be able to accommodate the different requirements for different languages (i.e. French or German language takes more room for saying the same thing)
- for example Manitobia.ca, the interface is available in French and English; and the suffix on the URL is /en/ or /fr/, indicating which version is open
Q: Is the Evergreen, or Open Source, solution also promoting the elimination of vendors?
*no, vendor relationships are still a good thing and have a place
*but the more open access relationships that can be build in the developing world, the more opportunity there is for a similar kind of library-vendor relationship and infrastructure to develop once the training is in place
- for example Project BlackLight is based on CNet which is a for-profit website, but because the code is open source, Project Blacklight adapted it to the library catalogue and user needs
* in terms of support, though, the benefit of the OS community is that there is generally more interest in supporting users compared to vendors who have already got you in their pocket, and consider their products “delivered”
*open source developers are really interested in making sure their products work
Q: How do initiatives like eIFL implement their technology in developing countries where political and infrastructural barriers exist that may not support OS initiatives or ethics?
* Zimbabwe are not democracies but Zimbabwe is still one of the biggest clients for eIFL
*eIFL works according to A. Maslow’s Hierarchy of Needs; in referring to this, Sadler, states, she does not mean “his ideas about motivation, [but] that you can’t think about library infrastructure in places where basic nutrition and safety needs are not being met” (Sadler, personal communication**, 2007)
- there has to be some kind of infrastructure, even if it’s legacy stuff
- there is a bar set for making a level of success
*bottom line requirements for pilot projects are in place because otherwise the project can’t meet its goals
- there needs to be enough people on staff in the host country who speak English
- host country need to be able to submit a certain amount of staff and hardware, need to have an automated system, i.e. an electronic catalogue
*eIFL can’t go in and solve all the countries’ problems at once, this is a niche service so it has to be implemented with realistic requirements and expectations
[notes by Jess Posgate]
**Sadler, personal communication, follows in comments posted belowAdvertisements