What makes me sad is that both Ackerman [sic] and Wallis have missed a key point: if the future is in web services, how can libraries take advantage of that with their current staff configurations? How many libraries in the U.S. have a honest-to-goodness computer programmer on staff? How many have staff with Computer Science degrees? How many staff do they have devoted to the library's hardware, software, and network? How many staff do they have devoted to web services?
In the smallest libraries, perhaps all of these are the same one person.
My article is about modern software engineering for libraries, not on how to staff them. However, in the very same issue of netConnect there's an article by Karen Coombs "Digital Promise and Peril" calling for library staffing to reflect the digital content environment
Many of these digital materials are in jeopardy of being lost because librarians have not yet found adequate ways to collect and manage them. In part, this is because roles and skill sets have been siloed in libraries. Materials preservation issues have typically been the purview of special collections and archives units within the library. In contrast, cataloging expertise has resided in technical services, and technology expertise has typically resided in systems. To collect and manage born digital objects adequately requires these roles and skill sets to come together.
So let me summarize some of the goals and targets of the article, as well as talk about the relationship to promising developments:
- The main focus of the article is to convey to library administrators, managers and planners that the world of networked digital content requires new ways of thinking about developing library systems, and that there are modern software engineering methods and technologies that can support new systems development.
- It may be the case that, as in Sarah's words, the library blogosphere knows that "Of course the future is in web services" but it took years of hard lobbying and education within my organisation to convince my own library leadership - that's why I was happy to have the forum of Library Journal, to reach a wider audience. I have to believe there are many library managers who have no idea where the technology future lies, and have only a vague notion of web services.
- If I can get one library manager to be able to ask good questions about web services and service-oriented architecture, and even (dare I dream) read one of the books I recommended with more detailed information on the topic, my goal is achieved.
- I recognize that many public libraries don't have the staff for development, and never will. That's why I have said in the past that librarians should be scripters, not coders. Most libraries should be using technology developed externally, not trying to do their own custom internal development, or as I said in my article, they should be technology service consumers, not necessarily producers. There are only a few libraries in the world (including mine) which have the size to have a substantial development staff.
- I work at a research library, not a public library nor an academic library. Research libraries like mine now interact with patrons almost entirely online - our walkin traffic is basically zero. That means we have to move into the networked information space very aggressively, or basically disappear as a presence for our users, replaced by publisher web sites.
- By moving to standard APIs, using standard modern development methods, and standardizing web services, libraries can take advantage of a much broader development community, and potential staffing pool. How many software developers do you think know Z39.50, MARC, and SRU/SRW? That's why we have a tiny community of library hackers trying to make things work. Now imagine that instead your library software job poster just says "developer needs to work with standard Java tools to develop software using modern methods for standardized APIs". Getting access to a better, broader staff base is intimately connected to moving library technology into the mainstream of software development.
Where can we look?
Maybe the DLF project on ILS APIs will help.
Maybe the OASIS effort on standardising search services will be useful.
Maybe it happens by using OpenSearch and simple REST interfaces rather than custom library protocols.
UPDATE: There is another important piece, which is about libraries reaching out and speaking the right language. That's why you need to understand how to express things in terms of SOA, Web Services, and APIs. There is way more innovation capacity outside your walls than you can ever get inside, even if you have the perfect IT staffing policy and budget. From your local superpatrons, highschool CS students, and local college and university computer science departments, to, basically, every CS student in the entire world. You can reach them with contests, with collaboration requests, with invitations to improve your systems... but here's the important part... if you speak their language. CS people love challenges and programming, but they're not going to learn obscure library jargon and usage like OPAC, Z39.50 and database (which means something completely different in CS). You can't say "hey, can you help us improve our OPAC because the Z39.50 doesn't federate across our databases". They're not going to know what the f*** you're asking. Learn the CS language, and a whole world of programmers will open up to you.
To me one of the single biggest missed opportunities is in the digital library community. Ever year, lots of computer science groups, flush with energetic grad students, toil away and produce results that are presented at JCDL and ECDL. And, based on my experience at ECDL 2006, they then present those results entirely to a community of other computer scientists. Where are the librarians? Why are you all at library conferences talking to other librarians? Come to *CDL and ask the computer scientists to build stuff you need. Yes, it's a difficult transition from research to production, but at least join the conversation. ENDUPDATE
The good news is that there are lots of projects out there already - I don't think it's a case that there is no activity. The fundamental point of my article is that for these projects, we have to use enterprise architecture, service-oriented architecture, web services / standard APIs and the whole toolkit of modern network-based standard-data software development. Because if we don't, WE WILL BUILD SILO SYSTEMS AGAIN.
[Sidebar on Scribd: be careful browsing around this document hosting site. Many of the profiles, profile images, and documents are unfortunately very not safe for work. Scribd really needs to put in a moderation / adult content filtering system.]
Why do I think it's important to talk about these topics? Because there really are lots of new developments in the library catalogue and OPAC world, including:
- Scriblio - "free, open source CMS and OPAC with faceted searching and browsing features based on WordPress"
- VuFind - "The goal of VuFind is to enable your users to search and browse through all of your library's resources by replacing the traditional OPAC"
- Evergreen Open ILS - including British Columbia Pines project - "The phased implementation of the Evergreen Open ILS for all of BC ("BC PINES") be implemented over the next 5 years. We hope that eventually all public libraries in BC will join"
- eXtensible Catalog (XC) - "an open-source online system that will unify access to traditional and digital library resources"
- CERN systems redevelopment - build a complete high-energy physics (HEP) information system with full-text, data-mining and demonstrate and deploy Web 2.0 applications in the domain of sciences
There are also some great modular browser tools out there, including
- LibX - a Firefox extension that provides direct access to your library's resources [and] an open source framework from which editions for specific libraries can be built.
- Zotero - a free, easy-to-use Firefox extension to help you collect, manage, and cite your research sources
I'm very much hoping that these developments will open libraries up to be better network participants, with a broader community of developers able to build pieces, and with standards enabling libraries with limited development capability to simply plug-and-play.
June 29, 2007 Casey Bisson on Scriblio and OpenLibrary