By Donald T. Hawkins (Freelance Conference Blogger and Editor) <[email protected]>
The Electronic Resources & Libraries (ER&L) Conference returned once again to the AT&T Executive Conference Center in Austin, TX on March 4-7, 2018. It drew over 1,000 attendees from 17 countries, as well as about 500 online attendees. Besides the traditional mix of plenary and concurrent sessions, the conference featured an exhibit hall with over 80 exhibitors, a number of poster presentations, and pre- and post-conference workshops.
Sandy (L) and Bonnie Tijerina, Conference Coordinators
Content Standards and Their Consequences
Robyn Caplan, Researcher at Data & Society Research Institute (https://datasociety.net/), noted that the power of platforms lies in the central position that search engines occupy. However, distinctions between media companies and platforms have begun to blur, leading to ambiguity in our perceptions of how people are consuming information:
- Platforms might be thought of as publishers, but large ones like Facebook, Twitter, and Google cannot define themselves well because it is difficult for them to categorize their content well.
- Are platform companies really media companies? Mark Zuckerberg, founder of Facebook, has said that Facebook defines itself as a technology company because media companies are largely known by the content they create.
- Are media companies platform companies? Some of them think of print as just another platform, and news media have shifted their strategies to be platforms.
- Barriers between distribution channels and publishers are converging—analytics are being used interchangeably. Platform operators are wondering what content they should be prioritizing, which are decisions publishers used to make.
- The internet was intended to break down barriers between producers and readers and was initially advocated as the true public sphere. But it is unlikely that early internet companies could foresee today’s situation that there would be only a few companies dominating the internet.
Caplan referred to a report recently published by her organization, which she co-authored: “Dead Reckoning: Navigating Content Moderation After “Fake News”. It discusses fake news as an illustration of the problems of categorizing online content. Fake news can be detected by
- Intent: the author intends to spread false information, and
- The type of information being conveyed, such as hoaxes, conspiracies, satire, etc., and
- Its features: unique words, visual cues, sensational images, and social patterns.
When we make content decisions, we are implicitly classifying some of it as good and some as bad. For example, in the fake news case, it is difficult to tell the difference between someone intending to inform and someone intending to deceive. Solutions to the fake news problem depend on:
- Trust and verification by third-party fact checkers, who must exist globally to be effective,
- Demonetization of content not meeting standards,
- Banning accounts known to be producers of fake news and de-prioritizing their content, and
- Regulatory solutions (generally used outside the U.S.).
Platforms now rely more than ever on people to moderate and categorize content types. Content decisions depend on context and involve difficult decision-making. Platforms are trying to balance consistent policies with flexibility, global platforms observing local laws, transparency, and protecting user privacy. They have vastly different resources for these operations, and are trying to distinguish themselves from each other.
Community Engagement: Using Linked Data to Increase Event and Collection Discoverability
According to Ravi Singh, Executive Director, Demco Software (https://www.demco.com/software), linked data creates visibility for libraries. He cited a recent Pew Research report, which reported that 80% of the adult respondents to a survey said that libraries should offer programs to teach people how to use digital tools such as 3-D printers, and 50% said that libraries should buy the tools. These data show that the needs of libraries’ communities are changing; libraries have responded by offering 3D printing, equipment for programming virtual reality (VR) applications, and empowering users to connect with their mobile apps. However, the same report also found that many Americans do not know that libraries offer learning-related programs.
Linked data opens up internal data to the web and encourages companies and others to publish it freely, but in libraries, linked data has been largely applied to books. Search engines use linked data to enrich their results, which can help libraries transform themselves from passive repositories into educational centers. Linked data works well with unique content and can drive awareness of events and resources because people expect search engines to deliver the information most relevant to them at the top of the results list. Libraries and communities thus benefit; Demco’s product, Demco Discover, provides a suite of tools to help libraries use linked data effectively.
How Do Students Do Research?
Molly Beisler, Discovery Services Librarian at the University of Nevada-Reno, presented a fascinating talk on her research into how students do their research. Instead of using the traditional method of having the searchers describe what they were doing, she had 222 students from eight classes draw diagrams. Images allowed expressions of ideas or feelings that might not emerge through words alone and also encourage abstract thinking. Some of her findings were:
- 89% of the images showed research, and it was good to see that some students mentioned the library’s databases.
- Students are using the library’s resources, and several of them mentioned the discovery system used at the library (Summon) by name.
- Help was used at various points in searches, and it came from multiple sources—peers, family, and the writing center. The preferred source of help was peers; it was not frequently sought from the library.
Here are some of the drawings made by the students.
I Have Never Seen Anything Like This: Student Interpretations of Metadata
Three researchers from the University of Arizona libraries reported on a study of 26 undergraduates in social and behavioral science programs which asked how students interpret metadata. The students were given some metadata terms from search results and asked to identify the subject of the articles, the producer, and whether the articles were relevant to the search topic. Here are some of the students’ reactions:
- “This looks like something I would not look at. I have never seen anything like this. If I saw this, I would go somewhere else.”
- “This looks like a handbook. I’m not sure if that is just a smaller book?”
- “Is this a PDF, book, or article?”
- “I look at the title, then subject, then creators or publishers to see if it is a credible source.”
Metadata is clearly not what students are used to seeing in their search results, even though it is the interface between the user interface and information literacy.
Using Personas to Meet Users’ Research and Scholarship Needs
According to Melissa Gustafson, Electronic Resources Librarian, Indiana State University, personas were introduced in the library world in 2003. They are research-based user types that can be used to highlight new opportunities, uncover gaps, and develop the functionality of a resource or service. Three to five personas are useful in a research project; qualitative personas are used in interviews, usability testing, focus groups, etc.; and quantitative personas are used in surveys and data analysis. After using the personas and analyzing the data obtained, resources can be refined and redesigned.
Listening to Stories: Partnerships Between Faculty, Publishers, and the Library
Lou Palmer and Bekah Shaw from SAGE Research noted that there is often a tension between faculty members and librarians, which is an obstacle to meeting library users’ needs, despite the role of the faculty in selecting content (especially videos) for the library. SAGE wanted to facilitate conversations between publishers, faculty, and libraries, and librarians were receptive to getting feedback from the faculty. So a “Field Editor” positon was created to work with academic departments and determine the best way to help the library make decisions. The result was that SAGE received information from libraries that was useful for product development.
Best Practices for Licensing Online Video Efficiently and Effectively
Erin DeWitt Miller and Andrew Trantham from the Media Library at the University of North Texas (UNT) noted that online video is fairly new, but it has become hard to imagine a world without it. Library users have come to expect it, and by 2020, it is estimated that online video on-demand viewing will be the equivalent of circulating 7.2 billion DVDs per month. As older media becomes obsolete, preservation aspects of online video become important, and libraries must be ready to support online courses using video.
Managing online video is complicated because there is no industry standard. Licenses for online video must accommodate demand-driven acquisitions, curated collections, and licensing by titles. By its nature, video in libraries involves copying (file digitization) and distribution (making the material available to library users).
At UNT, all legacy print licenses have been digitized and are stored on a departmental shared drive, so they are available to staff members and can be keyword searched. Here is some general advice regarding licenses:
- Clarity and specificity in the language are preferable.
- Maintain a clear understanding of your institution’s legal requirements and a strong working relationship with the legal counsel.
- Establish clear documentation policies and procedures.
- If you have specific needs, don’t be afraid to ask the vendor for them.
- Stay organized and track license expiration dates. Library users should be able to see all license terms.
- Plan for growth in budgeting and know your vendors.
As this slide shows, there are many steps involving significant time in acquiring online video after it is requested by a faculty member; it is important to recognize this in planning.
Digital Publishing: A Home for Faculty in the Library
Clare DeMarco and Kyle Courtney from the Harvard Law School (HLS) said that we are all doing digital scholarship, but we may not be recognized for it. Are these activities scholarship or publishing? Does it matter? We are dealing with digital content and have an opportunity to deepen the faculty-library relationship by showing the value that we can bring to the work of the faculty.
We must consider:
- What is the role of digital publishing in the library?
- Where can libraries look for support to remain competitive in the digital space?
- How can libraries support and develop faculty members who are creating digital scholarly content?
The faculty looks for customization, which is an opportunity for librarians. A library acting as a publisher may disrupt its traditional relationship with the faculty. Here are some examples of successful faculty-librarian partnerships at HLS:
- A faculty member wanted to start an open access online journal but did not have a funding source. It was difficult to get a URL and an ISSN for the journal. The library was able to help and provide these services.
- With textbook prices increasing at twice the rate of inflation, HLS developed an online textbook platform, digitized 42,000 pages of US case law, and made them freely available. Librarians must think about platform development and be assertive that they understand taxonomies, classification, etc.
- Librarians can help in developing proofs of concept for reference and bibliographic materials for general audiences.
These are examples of areas that we can enter now, and we should do it. Librarians must continue to demonstrate their value to the faculty and leverage their strength as a community. We should not be afraid to call ourselves digital content creators.
The Importance of Discovery
Kari Paulson, VP of Market Development, and Jed Reinitz, Director of Product Management at ProQuest, discussed how prioritizing exploration of content promotes the value of the library. If we acquire and index content, will our users find it? We need to think about how we are doing at multiple levels of discovery because discovery and content are intertwined.
Discovery is the ability for users to find the content they are looking for, or in some cases, content they did not know that they needed. It is a shared goal that all stakeholders must work together to achieve. Priorities are different for everyone:
- Users: connecting to content.
- Publishers: increase the visibility of their brand and content.
- Aggregators: optimization over multiple content providers.
- Librarians: connect users to the library’s content and demonstrate impact and usage of their resources.
The results of a study available from Simon Inger Consulting (now Renew Publishing Consultants, http://renewpublishingconsultants.com/) on how readers discover content in scholarly publications are encouraging because they show that many users still consider the library their most important resource. Twitter is now playing an increasingly important role. Most researchers find e-book content from within the book and rarely do further searches.
Aggregators provide metadata enrichment, indexing across many suppliers, holdings updates, keeping everything in sync, and platform searching. We need to be reminded of their impact when we get discovery right or when problems arise.
By bringing bibliographic information closer to users’ workflows, libraries can enable them to better evaluate available resources, create new pathways to serendipitous exploration of content (i.e. bringing back the experience of perusing the stacks), and provide a modern user experience that increases content usage. There are numerous pathways for discovery, and the user experience can be enriched with features available on other systems outside of the library, such as “more like this”, “about the author”, and “you may also like”, all related to the library’s resources.
Special Session: The Messy Reality of Algorithmic Culture
danah boyd, Founder and President, Data & Society; Principal Researcher at Microsoft; and Visiting Professor at New York University, began this special session by wondering what has happened to our algorithmic culture. She said that AI and Big Data seem to depend on the myth that if we just collect more data, we can solve the world’s problems. We are not dealing with technology but an image of social problems. For example, we can now see into the lives of people through social media platforms, so we must be able to deal with things more intelligently. The important point is not just the fact that we have the data, but how we make sense of it, which can be broken down into these four steps:
- How did we get the data?
- By choice: you know you want the data and get something in return, then share it.
- By coercion (the opposite of choice). There is coercion at multiple levels; people in your network pull the data out of you.
- By circumstance: you share something and hope something bad won’t happen.
- Seeing patterns, creating problems.
- We make sense of what we are seeing by discrimination: the action of perceiving, noting, or making a distinction between things; the power of observing differences accurately; treatment of goods, etc. on a more or less favorable basis according to circumstances; or unjust of prejudicial treatment of a person or group.
- How power comes into this is where the biggest problems lie. We may take our own discriminatory logic to the data, and how we use it can be an abuse of power.
- When do we use data? Do we make wrong assumptions? How do we deal with resulting recommendations?
- Explicit corruption of data.
- For example, brand names inserted into updates are more likely to be seen in a news feed.
- White text on a white background is invisible and can be used to game a system.
- Someone’s name on Google can be used to make an unfavorable political point.
- Hacking the attention economy has become fun for a lot of young people.
- We are living in an ecosystem where data is being systematically manipulated for a variety of purposes.
- Address values and cultural norms first.
- Our responsibility is to ask the hard questions first, learn from the data, and help people. Part of the process of analyzing information is prediction of the future. (For example, airline pilots are expected to step in when something goes wrong; otherwise planes are flown on autopilot.)
- Much of what we do could be smarter. We need to figure out what we want our tools to do. No technology is ever neutral.
Vendor Relations: Evolving Ethos and Etiquette
Laurel Sammonds Crawford, Head of Collection Development, and Allyson Rodriguez, Coordinator of Electronic Resources, University of North Texas Libraries, noted that today’s environment is changing significantly: demand for electronic information is increasing; budgets are being restricted; content is becoming increasingly complex; and vendors are consolidating. This process is rapid, ongoing, and difficult to track, which is putting pressure on both librarians and vendor representatives.
Librarians need to decide what they want from their relationships with vendors. The most important thing is that the relationship is a business transaction, and the financial implications are always there. Any relationship should be a means to an end, and we must always keep our users in mind: students have paid fees to attend the university, part of which goes to the library.
Negotiations are necessary and can be burdensome. Inevitably, conflicts with vendors will arise. We must always be professional, but that should not prevent us from firmly advocating for the user. Timely and accurate invoicing is important, and we must evaluate resources and decide if it makes sense to purchase them for our organization. Here are some tips for successful negotiations:
- Be prepared, stand firm, and be patient.
- Know who you will be negotiating with. A good attitude to take is “Here are my needs; help me meet them.”
- Learn and document everything in writing.
- Give vendors context about your needs and provide feedback.
- If you experience high pressure tactics, make sure to take notes on them and be prepared for the next negotiation session.
- Set healthy goals and boundaries. Think about your priorities.
- Guard your time. Limit the number of meetings; set a time limit and stick to it; require email confirmations of meetings.
Following these principles will result in strengthened relationships, focused discussions, and the best use of time. You have a large responsibility to make sure your users are getting the best information for their needs.
Going Where Our Users Are: Enhancing Discovery and Access
Michael Levine-Clark, Dean and Director, University of Denver Libraries, and Jason Chabak, Director of Institutional Sales, ReadCube (https://www.readcube.com/), described a collaboration and pilot project to improve access to content for users. Users now expect instant access to everything; often, after several clicks, they discover for a variety of reasons that they do not have access to an article: they must be on campus and know how to set up access; discovery tools do not point to open access versions of articles; access is at the journal level, not the article level. For many users, the library is hard to use, so they do not start their searches with the library, but instead, go to sources like Sci-Hub.
The University of Denver Libraries began collaborating with ReadCube, which has a history of working with publishers, researchers, and libraries. It provides a seamless user interface and integrated access to PDF references from Google Scholar, PubMed, and other search engines. Users can read articles directly while browsing the web and create their own personal library. The main pain point for users is access to articles; inconsistent and frustrating experiences are driving users elsewhere because of:
- Unique researcher workflows,
- Varying discovery tools and vendor authentication methods,
- Inconsistent metadata from publishers,
- Dead ends and broken links, and
- Difficult mobile access.
ReadCube’s vision is to provide access from anywhere that is easier than using rogue sites like Sci-Hub, so that users can easily get the content they need when they need it, libraries a tool that eases and enhances usage and reduces redundant purchases and downloads, and publishers will enjoy higher usage. So far, ReadCube has been focusing on increasing its coverage of scholarly subscription and open access journals. It expects to expand into additional formats such as ebooks, videos, and images.
Moving Beyond IP Authentication: The New Frontier in Single Sign-On
Andrew Nagy, Director of SaaS Innovation, EBSCO; Robert Scaysbrook, Head of Sales, OpenAthens; and Dale Saenz, Library Director, Laredo Community College, described how EBSCO has partnered with other companies to help libraries be as successful as possible. The typical end user journey is shown here:
But it can also be represented in terms of providers and their services:
There are many scenarios for accessing content; how can we streamline them and make it easy? Is there anything we can do to make it more of a managed service? Authentication is a significant problem in the user experience, so EBSCO has partnered with OpenAthens (https://openathens.org/) to create a fully integrated, librarian administered, single sign-on solution for libraries; even when the library is closed.
It is widely recognized that authentication based on IP address no longer works in today’s distributed world because access is increasingly through mobile devices; personalization is expected; and changing license requirements are resulting in more to manage. Multiple technology services must be integrated, and usage of collections must be monitored.
OpenAthens is different from IP/proxy authentication because it was designed by librarians for librarians and works across all devices and locations. It is a cloud-based service that collects detailed usage statistics, supports user personalization, and uses a security protocol (Security Assertion Markup Language (SAML)) to pass information about users to service providers. Using the OpenAthens system, EBSCO built tools for access, searching, and training users, and aiding libraries in their support for higher education. Savings in costs and time are the result.
The Bookstore is for T-Shirts: Cooperatively Marketing the Library’s Ebooks as Textbooks
Maura Diamond, Director, Institutional Sales, Springer Nature; Kelly Robinson, Head, Collection Management, Embry-Riddle Aeronautical University (ERAU); and Krystie Wilfong, Associate Librarian for Collection Management and Scholarly Communications, Bates College, addressed the widespread problem of high textbook costs for students. Robinson reported on a survey of 22,000 college students in Florida that concluded that the high cost of textbooks (averaging approximately $1,200/year) is negatively affecting students’ success and course completion. About 2/3 of the survey respondents said they did not purchase a required textbook, and over half got a poor grade or failed a course as a result. Alternatives to textbooks include:
- Open Educational Resources (OERs) curated by the library,
- E-journals permitting fair use, and
- DRM-free e-book collections.
Advantages of these alternatives are used in marketing them to faculty, some of whom were worried about funding, effects of OERs on promotion and tenure prospects, and the desire to have control over which textbooks they use in their classes.
ERAU is experiencing retention issues and pressure from the state to lower costs. It is difficult for their faculty members to find textbooks for their courses, so the library has created reading lists of them which were deposited in their institutional repository. These initiatives are growing, and more library books are being used as textbooks.
Willfong noted that Bates College’s reserve policy has changed since 2013. Formerly, traditional textbooks were not included, but now the library has a policy that all required textbooks are put on reserve. They have arranged for the library bookstore to proactively promote reserved books to students. Although the average use per item is 10 copies, some books have been accessed over 200 times. And sales of those books by the bookstore have significantly decreased. The faculty is in favor of these efforts because all students have an edition of the required books.
Maura Diamond said that over 6,900 librarians and professors have expressed interest in Springer Nature’s “Affordable Textbook” campaign, in which they used the library’s records to identify which books were being used in courses at Hope Colleges. Editorial Board members of book series are working with Springer Nature to develop materials that they can use to market licensed ebooks in their departments. After the appropriate books were identified, a toolkit was created for librarians to use to find which textbooks are available and create awareness with students.
Antisocial Media: How Facebook Disconnects People and Undermines Democracy
Siva Vaidhyanathan, Professor of Media Studies at the University of Virginia, closed the conference with a revealing look at Facebook and how it controls many aspects of its users’ lives. Much of his address was taken from his forthcoming book Antisocial Media, which will be published in September 2018. He began by marveling that Facebook now has over 2.2 billion users, and this number is rapidly growing, especially in Myanmar, Brazil, and India. Facebook’s engineers must therefore deal with a vast universe dispersed around the globe—it has become a system too big to govern.
Vaidhyanathan’s main premise is that if you wanted to invent a propaganda system to promote nationalism and authoritarianism, you could not invent one better than Facebook. It scrambles our social, economic, and political contexts. All of its users are in the same social milieu, which is not the way we live; we manage our lives and reputations according to the relationships where we are at the moment; which is how we are social. In contrast, Facebook wants us to interact with as many people as possible in the same way because that is in their interests; its developers believe that the more people we share things with, the better our lives will be. We are unable to have collective calm deliberations about anything important because our news feeds are constantly interrupting us and capturing our attention. Vaidhyanathan said that is a disaster when it is expanded to a universe of 2.2 billion people.
Facebook works very well at doing what it is designed for: creating addictive products that present ads to us in more aggressive and effective ways than ever before. It has a huge breadth of influence in the world and controls the top 5 social media platforms (outside of the restricted Chinese systems). It scrambles our relationships in three major areas:
- Personal. Facebook accelerates the distribution of content that is likely to generate clicks, links, and shares (for example, cute puppy, cat, and baby pictures). This type of material appeals to our emotions, even negative ones, such as condemnation of hate speech, and generates many strong comments on both sides of an issue.
- Commercial. Every marginal dollar spent on advertising in the last two years is spent on Facebook. Advertising used to be a faith-based practice; there is no way to accurately measure the impact of a print ad, but now with Facebook’s system, it is easy to target ads to a precisely defined audience, measure their effect, and alter them as necessary in real time. Marketing budgets are therefore spent much more effectively; it is not uncommon to run an ad on Facebook that cost in the hundreds of dollars but results in thousands of dollars in sales.
- Political. Facebook’s internal investigation of Russian political ads in the 2016 election showed that they undermined faith in the US democracy. Other examples are well known: Donald Trump’s campaign officials said that he won the election because of Facebook’s ad targeting; Narendra Modi, the current Prime Minister of India, runs all of his political campaigns on Facebook and uses it to threaten dissenters; and corrupt elections in various countries have been traced to Facebook ads. There is no oversight, transparency, or opportunity to respond to such actions.
Vaidhyanathan’s conclusion is that the reason Facebook’s mission went wrong is that its leaders believed that good intentions were enough, and that blind faith in technology could generate a better world.
For a very general overview of the conference, see “Top 10 Trends at the 2018 ER&L Conference,” https://hub.wiley.com/community/exchanges/discover/blog/2018/03/18/top-10-trends-at-the-2018-electronic-resources-libraries-erl-conference. The 2019 ER&L conference will return to Austin on March 3-6.
Donald T. Hawkins is an information industry freelance writer based in Pennsylvania. In addition to blogging and writing about conferences for Against the Grain, he blogs the Computers in Libraries and Internet Librarian conferences for Information Today, Inc. (ITI) and maintains the Conference Calendar on the ITI website (http://www.infotoday.com/calendar.asp). He is the Editor of Personal Archiving: Preserving Our Digital Heritage, (Information Today, 2013) and Co-Editor of Public Knowledge: Access and Benefits (Information Today, 2016). He holds a Ph.D. degree from the University of California, Berkeley and has worked in the online information industry for over 45 years.