By Donald T. Hawkins (Freelance Editor and Conference Blogger)
Attending der Frankfurter Buchmesse (that’s German for “The Frankfurt Book Fair”) is an overwhelming experience but one that every information professional should have at least once. The fair is generally held in mid-October and attracts several thousand users to a week-long extravaganza of vendor exhibits in several huge halls. It is the publishing industry’s largest trade fair. Because of the COVID-19 pandemic, the 2020 Book Fair was held digitally. Here are some statistics:
- 200,000 users on buchmesse.de
- 4,440 exhibitors from 103 countries
- 3,644 events in the Calendar of Events
One feature of the Fair is a concurrently running conference featuring seminars on a wide variety of topics related to the publishing and information industries. This year, the Society for Scholarly Publication (SSP) and the Charleston Conference participated in the conference; this article summarizes both of them.
The SSP seminar was moderated by Angela Cochrane, SSP Past-President. It featured discussions by four of the “chefs” from the Scholarly Kitchen on the topic “Building Back to More of the Core”.
The goal of this seminar was to examine the way forward from the COVID-19 pandemic. What will remain the same, and what will have to change? In her introduction, Angela Cochrane said that we must figure out how to keep operations going and make sure that users can access content they paid for, and that critical research is being reviewed and made accessible as quickly as possible. What will the “new normal” look like, and how do we get there? We don’t want to go back to “normal”; we want to go back better. We are in this for the long haul; full financial recovery may not occur until 2022 or 2023.
Todd Carpenter said that not all institutions were set up to deliver remote content and asked what we have learned about access and what can we make better going forward. Authentication systems for off-campus users have shown considerable strain because many proxy systems could not accommodate the bandwidth requirements of so many users. The one thing that publishers can do for libraries is to adopt single sign on to alleviate burdens on proxy servers. We have seen a dramatic spike in federated identity authentication by publishers, particularly those who had implemented a seamless access service. As institutions, especially libraries, closed they had to figure out how to continue delivering traditional forms, (mostly physical items) of content.
Another initiative that has exploded has been the approach to controlled digital lending (CDL), as a stopgap method of fulfilling requests in a digital environment. The Internet Archive (IA) was an early vocal supporter of CDL, but other institutions are also seeking safe and secure approaches to circulating items in a safe and secure digital manner. There is now a lawsuit against the IA over this approach. It might be more gracious for publishers to allow libraries to do CDL for a limited time under limited circumstances.
Lisa Hinchliffe said that although some large institutions have long had remote access to their platforms, the infrastructures of many smaller ones were stressed as all of their content moved to remote access. For example, the University of Illinois subscribes to over 1,000 databases. They are not all able to function with remote access, so we must figure out how to provide users with the institutional password.
Complicating this issue is that many publishers generously opened up their platforms to end users when COVID occurred, but the result of which was that we do not know what the users are doing during the pandemic, or if they are even from our institution. That generosity has now ended, which has resulted in an immense amount of extra work for libraries to restore the previous arrangements. Another problem is that many users have discovered piracy sites and are happy with them because all they have to do to get a PDF is enter the DOI of the desired article. (Users want the DOI to PDF route to be as short as possible which is not an experience they can get using a publisher’s site.)
The pandemic has exposed the absolute failure of our e-book delivery systems and pricing models. Libraries are having difficulty finding copies of e-books to buy because of publishers’ decisions to not sell to libraries. We have seen some incredible price gouging and unsavory behavior during this time. E-books are broken!
Tim Vines discussed data access and said that the unit of research should be the journal article. When trying to do a meta-analysis of a group of articles, it is much more effective to use their underlying data sets rather than trying to reanalyze the data in the articles. We must improve our data sharing, particularly with biomedical data. Some researchers have suggested that it would be better to use funding grants as the units of research, but that causes problems because a single research project might be funded by several different grants. Focusing on articles allows us to ensure that all the data from a project are available. The challenge is figuring out which data sets to use in the analysis. Data sharing is not something that authors willingly want to do.
Sian Harris asked what we have learned about access-to-read or access-to-publish and whether the pandemic crisis has made such issues worse in developing countries. She said that there have been no noticeable differences in the need for access before COVID-19 occurred or afterwards because the need for access to research is continuous and does not switch on and off depending on external circumstances. Therefore, there are still gaps in access for lower income countries. The technology and infrastructure of what is available to researchers has changed because the pandemic has exacerbated existing inequalities. In countries like ours, we tend to take internet access for granted, but there is an enormous variability around the world in its cost, which is a big challenge.
We need to recognize that publishing is not done just in the US and the western world. There are publishers all over the world, and many of them have different models. There has been a massive increase of papers on COVID; many publishers have created special issues for them, and they play an important part in the global publishing system.
The Charleston Library Conference Meets Frankfurter Buchmesse
This seminar, moderated by Leah Hinds, Executive Director of the Charleston Library Conference, took place on October 16, 2020 as part of the virtual conference held along with the Frankfurt Book Fair. It featured presentations from seven information professionals.
Jim O’Donnell, University Librarian, Arizona State University discussed the state of the academic library. He said that we have moved into a new space in the last six months, and that the economic impacts of the COVID-19 pandemic will be with us for the next year and probably longer. O’Donnell identified two major trends now occurring in academic libraries:
- We have seen a new sharp move towards emphasizing the importance of digital content in the library. The preference for print disappears when we need information wherever we are. We have been slow in moving into the world of digital information and are now trying to have universal access to all information. However, we are at great risk of losing materials on the 20th century. We should remember and learn from it.
- How we acquire and use information is changing. Libraries used to collect things. We don’t do that anymore; but our models and how we think still emphasize the word “collections”. They will move to what we now call “special collections”. We now must find access to materials we do not own. Budgetary pressures now coming on us will combine with a flood of new information
The results of these trends will be:
- The pressure to move towards OA will increase, and
- Research universities are beginning to pay for the privilege of publishing material. We must find ways to use just-in-time models instead of just-in-case. This move must be in a different direction if we are to sustain our business models.
We have a great future which will be very different.
Charles Watkinson, Director, University of Michigan Press and Associate University Librarian, University of Michigan discussed the future of the monograph by describing several features of them.
- Today: Typically, monographs contain the results of deep investigations by scholars for scholars; however publishers commonly regard monographs as books that don’t sell. A transition to e-book sales is happening rapidly. 86,000 monographs are now published every year by over 8,000 academic book publishers. University presses regard themselves as guardian of the monograph.
- Tomorrow: Monographs will become a process as well as a product. Short form monographs of about 30,000 words are starting to appear, and they are sometimes about work in process. The University of Minnesota has built an open source platform for publishing monographs. Some monographs are “decoupled” books in which production is separated from distribution. They cost $11,000 which is less than the cost of a traditional monograph.
- Multimodal monographs are born “non-textual” and cost $8,000 to $12,000. Publishers cannot recoup the costs of many of them. Some are born audio or as smart e-books which have new challenges, such as distribution and preservation.
- Monographs as a social influencer: The need for authoritative information on COVID makes them essential. Many publishers have made their content free to read. There has been an extraordinary untapped demand for monographs.
Ivy Anderson, Director, Collection Development and Management Program, California Digital Library described how traditional approaches to collection development are changing and said that they are evolving from value-based to values-based. (Libraries have always been values-driven.) Much has been written about the user experience and focused on a dollar basis, but less about values. Traditional approaches have been quantitative. For an example, see the University of California library’s E-Book Values Statement.
Libraries have changed from being information acquirers to disseminators and have become enablers of the broadest possible use of information. See UC’s Declaration of Principles for Scholarly Communication. Here are the key values in collection development today:
MIT has developed a guide for negotiations with publishers, and over 100 institutions have endorsed it.
UC’s Office of Scholarly Communication has issued guidelines for OA agreements.
Jill Stover Heinze, formerly Director of User Experience at the University of Virginia and now a Founder of Saddle Stitch Marketing, LLC described developing and integrating a marketing program for the New York Public Library. Marketing plans have four elements: people, products, conversations, and iteration. Even when plans go awry, marketing must continue; change is happening all the time. We can see some good examples of this with COVID:
- Health, safety, and social support dominate concerns,
- Physical space and contact go from comforts to liabilities,
- People disperse and have varying access to technology, and
- Feedback loops, distribution and workflows are disrupted.
The entire value proposition must be reevaluated. What does it mean when
- The “library as a place” is a scary thought,
- To be inclusive when you must turn people away,
- To manage cultural resources when they must be behind locked doors, and
- To be closely embedded when social distance must be maintained?
We must put people and compassion first, realign product offerings with value, evaluate all points where users have contact with us, designate marketing communications as dialogs, and intentionally create and iterate a feedback loop. We don’t know the future, so we may not get everything right every time. Just because change is not sudden and dramatic does not mean it is not happening These elements should be incorporated into our libraries.
Anja Smit, University Librarian at Utrecht University said that copyright is at the top of academia’s agenda, but it has many complex digital issues. It requires collaboration with legal and other departments of the university. Students must know what is allowed when using works of other researchers, and researchers must know what use others will make of their work, both of which are governed by the local law of the country. Researchers are also teachers, and must know what is allowed in the classroom and how to disseminate publications. There is a growing pressure on libraries to check on materials used in teaching. Academic libraries are service organizations and are there to help the teachers. The best solution to copyright problems is to make sure that all fees are paid up front.
Todd Carpenter, Executive Director of NISO presented a review of AI (in 8 minutes!) and asked whether a Google search for AI will transform the future of communications. AI today is an algorithmic approach to problem solving powered by rapid replication of a process that is based on an explicit understanding of the desired outcome. It employs the brute force of strength of computer technology. A large part of AI is machine learning—figuring out how to solve predetermined conditions. Machines do not have a comprehension of anything we give them at the outset, and they have no biases. They cannot understand nuances, implications, etc. and do not think like people do. AI also has policy and legal barriers. It is not a creative approach to problem solving, but a logical one. Machines are not as smart as we might think: they are tools suited to do one thing and do it well.
To gain an understanding of the desired outcome, you must figure out how to achieve the goal. Machines are very good at memorizing. Higher levels of thinking and creative acts are not AI but are done by humans.
There is lots of opportunity for growth.
Carlo Scollo Lavizzari, an attorney specializing in copyright law in Basel, Switzerland discussed AI and publishing and said that we are in a tsunami of data more than ever before. Machine learning must be human assisted; as machines get better at one thing, they are not as good at others. Social sciences are becoming more important. AI is part of a fourth revolution in which machines will do logical things. Social interaction will be more relevant, but humans are important in setting things up.
Copyright adds a human dimension to AI. The future reader is an AI machine. Facts harvested illegally or without care are meaningless and unsafe even in the best AI machine. Licensing ensures the integrity of data inputs. The ethical dimension is most important. Leaving things unchanged can lead to more polarized outcomes. AI uses human-made intelligence and must work for humans. Content with AI becomes more versatile, but AI without content lacks direction. Here are some recommended next steps:
Roy Kaufman, Managing Director, Business Development at the Copyright Clearance Center concluded the seminar and presented the US approach to data mining. AI and text extraction are overlapping. We do not have a lot of law about text and data mining. Text and images are copyrightable; data is not. We must look at uses to consider Fair Use. The Berne 3-step test for Fair Use is useful.
We also must realize that exploitation today might not be exploitation tomorrow. Here are three illustrations from real cases:
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.