Technology continues to rapidly advance, often at such a pace that libraries struggle to keep up. The technical and financial capacity of the commercial sector far exceeds the resources available to any single library or even the largest vendors or development initiatives. However, once user expectations and technical standards are set on the broader web, libraries have an imperative to deliver their services accordingly. Five key trends in technology that impact libraries are attention to privacy, providing personalized services, enabling easy access to mobile devices, deploying services through cloudbased infrastructure, and the increased adoption of the microservices architecture. In this column, we will take a closer look at these trends and suggest their implications for libraries.
Privacy and Security
Although not a new concern, it has become even more urgent for libraries to deploy technologies that adequately protect the privacy of their patrons as they use the web-based services libraries provide. The library community has lagged behind the commercial arena in the implementation of technologies that are able to protect the exposure of online activities to untrusted third parties. These technologies include using HTTPS encryption to deter thirdparty eavesdropping on web sessions and avoiding the insertion of tracking tags that potentially connect a patron's session with advertising networks, which collect personalized information and contextual data.
The commercial sector almost ubiquitously implements HTTPS for encryption and follows rigorous practices to ensure the safety of customer details and financial information. Even the smallest lapse can expose vulnerabilities, enabling potentially catastrophic security breaches to their business. While these commercial entities deploy strong security, they do not necessarily do so to protect the privacy of those who use their services or visit their sites. Commercial web destinations aggressively collect, exploit, and monetize the activity of people as they use their web-based services. A person visiting Amazon or Facebook, for example, might not be as susceptible to an unknown third-party eavesdropping on his or her browsing activity. However, the owners of those sites will collect that information, and it will quickly proliferate into the advertising networks. Most of us have experienced the intrusion of ads displayed on one site after searching for related terms or products in an unrelated context.
Recent events make the imperative for libraries to deploy HTTPS on their web resources even more urgent. Web browsers increasingly flag any nonencrypted page as untrustworthy. The Google Chrome browser has progressively elevated the strength of its messaging relative to pages that fail to use HTTPS. The URL bar in the browser has already been adjusted to flag nonencrypted sites with an informational status warning that information viewed may not be private. Beginning in October 2017, Chrome (Build 62) began displaying explicit warnings that a site is not secure any time a user enters information on a non-encrypted page, including typing into a simple search box. Other browsers are likewise strengthening warnings about unsecure pages.
Libraries have been very slow to shift to HTTPS for their websites, online catalogs, discovery services, and other webbased services. Such sluggish adoption of readily available technologies stands in sharp contrast to the profession's stated value of protecting patron privacy. According to data collected in the libraries.org registry of libraries that I maintain, less than 10% of public libraries in the U.S. have implemented HTTPS (librarytechnology.org/libraries/ link-status.pl). It is ironic and embarrassing that libraries exhibit such an abysmal performance in implementing basic technologies that support their value of privacy. Many factors are converging to establish HTTPS as the expected protocol for accessing websites, amplifying privacy and security as key technology trends for libraries.
Key implication for libraries: Implement HTTPS for website, online catalog, and other web-based services as soon as possible.
Personal, Social Connections
Personalized services and social interactions have become expected components of web destinations. 1D content will not attract the same level of attention and engagement as sites that intelligently customize the experience of the user based on known personal interests and provide mechanisms to respond to, react to, or review content. Libraries have been striving to strengthen the personalization and engagement flavors of their websites and search services for more than a decade, with varying results. Early efforts included the MyLibrary concept that provided subject-specific links and content based on a patron's profile. The customized environment was activated for patrons once they logged in, which eventually proved to be a strong barrier.
The next-generation online catalog movement of the mid-2000s made progress toward providing a library search experience with a more modern interface, with features such as faceted browsing and relevancy-based ordering of results. The advancement of library services to incorporate a more personalized and social flavor continues. Bibli°Commons, for example, brings these concepts into the design of its discovery service and portal products, which have been adopted by a significant number of public libraries. Other public library catalog products are likewise evolving to sharpen their personalization features. In the academic sector, discovery services are advancing to deliver access to a more comprehensive representation of the universe of scholarly resources, and also provide tools that help researchers navigate to content that matches their interests as reflected in their personal profile or other contextual data.
Relative to general platforms such as Facebook, Twitter, and LinkedIn; academic networks such as ResearchGate; or discipline-specific communities, library platforms embody fairly weak forms of social interaction. As these products enter their next phase of development, I anticipate a stronger emphasis on personalized and social features.
Key implication for libraries: Implement services with enhanced personalization features and encourage their development via commercial and open source development initiatives.
Domination of Mobile
The level of access to web resources by mobile devices continues to increase relative to desktop or laptop computers. The general web passed the tipping point a few years ago; the proportions may not be quite the same for library resources. While it is unlikely that academic library users will perform in-depth research and writing on smartphones, many will conduct preliminary research, check sources, and conduct other activities. Many public library patrons are likely to access the library website, catalog, and services exclusively via their mobile devices. For public libraries, the ability to support ebook lending, downloads, and reading to smartphones and tablets has become a critical service.
The development of mobile-friendly interfaces and their implementation in library services have consistently fallen behind the general adoption and use levels of smartphones. Most of the major proprietary and open source online catalogs and discovery services have by now developed interfaces that follow a responsive design to accommodate mobile devices. Libraries tend to be slow to implement current software versions, meaning that these features may not be available to their patrons. Although I have not collected systematic statistics, as I visit library sites for updating entries in libraries.org, I see that a surprising proportion still are not mobile-friendly.
The relative immaturity of current interface designs and the sharply rising proportions of access will drive a new wave of development that emphasizes the excellence of user experience for those accessing library services via mobile devices.
Key implications for libraries: Ensure that all web-based services offered embody a responsive design, and consider deployment of mobile apps. Cloud Technologies
A dramatic shift is underway, moving from deploying library systems on equipment housed on-premises to some type of hosted infrastructure. Over the last decade, more libraries implementing new ILSs have opted to have them hosted by their vendor. Even those not interested in migrating to a new system will choose to shift to a vendor's hosting services when their existing server reaches the point at which it needs to be replaced. All new library automation products have been built specifically for deployment through cloud technologies. The genres of library services platforms and index-based discovery services have been designed for deployment through the cloud-based technical infrastructure hosted by their respective vendors.
This trend applies to almost all other categories of software or content platforms that are oriented to libraries. Institutional repositories, scheduling utilities, web CMSs, and subject-guide platforms are increasingly adopted via a subscription to a hosted service rather than installed on local servers. In most cases, cloud technologies provide more reliability, better performance, and stronger security than local implementations. They are usually deployed in industrial-strength data centers with more redundant equipment, monitoring, and engineering expertise than would be available in any given library that installs the software locally.
This trend fits well within the priorities of libraries, which have generally decreased the proportions of personnel dedicated to IT relative to other roles that more directly serve their patrons. The core competencies of libraries lie in the provision of information services and not in the implementation and maintenance of technical infrastructure.
Key implication for libraries: Evaluate the relative value of locally installed software as opportunities arise to deploy alternatives via subscriptions to hosted services.
The microservices architecture has emerged in the last few years as the preferred approach to the creation of complex and large-scale software applications. This architecture relies on the creation of many independent software components, each deployed on its own hardware platform, which work together to form a business application. Each microservice performs a narrow and well-defined task, maintains its own data stores, and is constructed with the internal components best-suited for its designated role within the broader application.
Microservices are usually deployed through containers, a virtualized infrastructure technology that makes efficient use of hardware resources. Containers have pre-defined configurations, which specify operating environments, software libraries, applications, and other internal components that enable very rapid deployment of services. Applications built with the microservices architecture can be scaled almost infinitely, as use increases, by launching multiple instances of each service. A microservicesbased application will usually include an API gateway and other components to route service requests and balance the load among the multiple instances of available services.
Microservices contrast with monolithic applications in which the entire codebase is compiled into a single set of executable files. The monolithic development approach has been the standard for most of the history of computer programming. Larger-scale business systems have increasingly adopted the services-oriented architecture, in which the software is based on many smaller units of functionality, although they're tightly integrated into a unified application. Microservices bring the concept of services-oriented architecture into a more distributed model, in which the components are developed and deployed independently. The internal architecture of a microservices-based application is not apparent externally since a unified interface manages the interactions with its internally distributed components.
Many of the large-scale tech companies now base their business infrastructure on microservices, including Uber, Netflix, Google, Amazon, and Comcast. Many of these companies originally deployed monolithic systems and faced the need to break their software into microservices to meet their requirements for rapid development and massive scalability. Until lately, the microservices architecture has not been commonly deployed in the creation of library-oriented applications. Even the recently developed library services platforms were made prior to the time that microservices were considered a mature development style. The open source FOLIO initiative was the first software in the library arena to specifically position itself as based on microservices. Bibli°Commons recently published a blog post discussing its technology stack, which includes a microservices component.
Key implication for libraries: The microservices architecture will increasingly be adopted into strategic library applications, such as resource management systems and discovery services. The internal architecture of applications should not necessarily be taken as a reason to adopt a given product, but more as an indicator of the possibilities for scalability, rapid development, and modular design.