Internet Librarian 2021, day 3 #InternetLibrarian

Accessing Knowledge: Internet Librarians’ Call to Action!

Jean-Claude Monney, Microsoft, Keeeb, and elsewhere; knowledge management

“Instant, relevant knowledge in context”

Knowledge discovery approaches:

Use a computer

  • Search: average 1.8 words in a search box; also people quit after 1-2 pages of results
  • Browse: not sure what you want (e.g., shopping)
  • Pull

Talk to a person: requires trust (“pedigree”)

  • Personal network
  • Professional network
  • Company network

When you get an answer in a document, what is its “pedigree”? Author, publisher, etc.

Knowledge is contextual:

  • Operational knowledge: knowledge is the end
  • Strategic knowledge: knowledge is the start

Transformational technologies:

  • Conversation as a platform
  • Cognitive services: AI, machine learning
  • Mixed reality

McKinsey (2012): 19% of time is lost in search
Microsoft (2019): 14% for executives, 11% for knowledge workers

Personal tech is signal rich (e.g., your phone tells you to leave early because of traffic)
Professional tech is signal poor

The shift from pull to push


  • Universal pull from everywheere
  • Conversational flow to find out what you want and how you want it
  • Automate push (AI, ML)

Open Access for Everyone

Carolyn Morris, Sirsi Dynix
Linda Barr, Austin Community College

Surveys of librarians show doubt and uncertainty about open access.

2020: 54% of articles sampled were open access. Many of them in journals that are not open-access only.
On trend for 100% by 2040

Funders (e.g., philanthropies) are driving open access. Also universities requiring that their faculty publish in open access and university libraries requiring it for their faculties as part of subsccription deals.

Major publishers are allowing open access (hybrid journals) to attract quality authors

Sirsi Dynix assists libraries in adding open access journals to their catalogs

Austin CC:

Collection development librarians didn’t feel they had the time or need to research scattered OA journals. However, there was an interest in open education resources (OER).

Did pilot of CloudSource, Sirsi Dynix’s product

Analysis of overlap of open access sources with paid resources

How do we evaluate OA?
How does it change how librarians work?
How should librarians think about it?
Is it all good, peer-reviewed?
Is the labor involved worthwhile?
It’s free. Should we take it all?
How to add it to library instruction?

They put the OA records in a separate database. The metadata varies and looks different.

Remote Working, Mixed Mode Working, or Hybrid?

Sophia Guevara, SLA, columnisst

Had to arrange your own technolgy and tech support. And furniture.

Meetings: video stream or not? If not, show a picture. Can have a virtual background or blurred background. [Note in chat says blur is disturbing to some people.] Consider lighting.

Mobile hotspot

Shared calendar (e.g., Outlook or Google Calendar)

E-mail, Zoom, MS Teams

Nimbus screenshot and screen video recorder

Krista Ford, Steptoe & Johnson

Pre-pandemic, almost everyone was in the office and had a learning curve to work from home.

People coming back to work. Different rules in different places for masks and vaccines. Also, differnt attitudes.

Modified work schedules, work-life balance. Could start early, work late, with a long break in the middle. Now, back to 8-5.

Lessons learned:

  • Remote work as an option
  • Workflow: assignments based on competency not bandwidth, use technology, eliminate redundancy
  • Targeted communication: don’t overwhelm people with e-mail and Zoom
  • Reduce print collection

Evolving technology:

  • Watch for new tech
  • Use old tech better
  • Train staff
  • Simplify: aggregate to limit access points for users
  • Use AI to drive strategic decisions

Redefining what it means to be an info professional:

  • Technology guru
  • Innovator
  • Information advisor
  • Social media influencer
  • Strategist

Creating Infrastructures for Long-Term Digital Preservation

Raymond Uzwyshyn, Texas State University

Long term = 10 years or more

Three-legged stool (Kenney and McGovern):

  • Organization
  • Technology
  • Resources

Any solution must allow for:

  • Diversity of tech
  • Replication (LOCKSS)
  • Digital auditing
  • Disaster planning (geographic distribution)
  • Best practices
  • Succession planning (for people and institutions)


  • Migration and preservation
  • Risk mitigation for data

Step 1: Form a digital preservation working group: technology and policies

Digital preservation policy

Archivematica: middleware standard for metadata and file integrity

Step 2: Conduct initial storage size estimate

Step 3: If possible, join consortia

Step 4: Storage infrastructure recommendation

Environmental scan of peer institutions
Narrow focus to three candidates. Classes: outsourced, in-house, mixed

Future directions:

  • Web archiving. Interests: Texas history and culture; university archives. Using Archive-It and Webrecorder
  • Digital forensics
  • E-mail archiving

Conclusion: a necessary focus area for research libraries.

AI Road Map for Academic Libraries

David King and Linda Kung, University of La Verne

Frank Lin, CSU San Bernardino

Use cases:

  • Analysis of search behavior
  • Chatbot/virtual assistant
  • Resource recommendations
  • Service recommendations

Other possibilities:

  • Content indexing
  • Document matching
  • Citation
  • Content summarization
  • Quality of service
  • Impact factor
  • Operational efficiency
  • Data analytics

Libraries’ Biggest Challenges & Opportunities for 2022+

Mary Ann Mavrinac, University of Rochester Libraries

Research libraries have been planning a digital future for decades.

Renovated spaces.
“Studio X”: AR lab area
Extending access to collections: Hathi Trust, Internet Archive
Reimaging education: open education resources (OER)
Library fellows
Connecting with community orgs.
Cultivating an inclusive climate
Wellness space in the library

Jim Peterson, Goodnight Memorial Library, Kentucky

Had to move to a warehouse building during pandemic
3-D printed PPE
Curbside pickup

Richard Huffine, FDIC Library

Well prepared to be virtual and still not in the office
MS Teams
Hybrid work: part in the office and part telecommuting
Many libraries still have physical collections and service desks
Preservation role (e.g., works on the history of our agency)
Balancing landline, cell phone, MS Teams
Rethink relationships with vendors

Hannah Byrd Little, The Webb School, Tennessee

7 library changes she hopes will remain post-pandemic

Online collaboration with faculty
Personal devices
Electronic collections
Reading for pleasure and personal growth (kids didn’t have any many other extracurricular activities)
Appreciation of time offline
Making online resources easier to use


  • Parents and governments getting involved in school curriculum
  • Learning gaps during pandemic
  • College and career preparation may need to change
  • Questions about future of work (e.g., 9-5 work schedule, which affects school schedules)

Ranganathan’s 5th law: “The library is a growing organism.”

Suzanne Marsalisi, Communico

Handling pandemic and post-pandemic worlds


Internet Librarian 2021, day 2 #InternetLibrarian

Publishing Models: Paywall vs. Public Access

Michelle Manafy, Digital Content Next, an advocacy group for news publishers

About half of Americans get their news from social media. Often the news we see is disintermediated — headlines, quotes, pictures divorced from context.

At the same time, trust in the news media has gone down.

Misinformation, disinformation, and malinformation. The last is deliberate fake news.

Algorithmic amplification of social harms. Facebook’s 2018 algorithm change actually rewarded outrage.

Google and Facebook eat up 80% of digital advertising.

In response, media companies try to get revenue from paywalls:

We need quality news. How do we pay for it?

Some publishers are allowing users to share an article.

Moderator: I get these industry newsletters, and I don’t even bother to click on the link.

Richard Huffine: I see article for free on aggregators like Yahoo News.

M. Manafy: Those sources are not very lucrative for the publishers. Apple News pays better.

Also, nonprofit news organizations are trying asking for donations (the NPR model).

Automatic writing for structured news like business and sports helps, but doesn’t fill the gap.

Michael Blackwell, director of St. Mary’s County Library (Md.)

Library digital content:

Some publishers won’t sell digital content to libraries at allowing

“Windowing”: limit to one copy per library (Macmillan)

Older content (before 2000) has often not been digitized

Siloing: content only available on one aggregator, but users get used to one and not others

License terms: With physical books, you own it and can circulate them one user at a time, often 50-60 circs over the license. E-books are often more expensive than the print equivalents. An e-book could cost $60 and be limited to a two-year term. It’s a much higher cost per use. This discourages libraries from buying new fiction, some nonfiction, etc. Some publishers offer a set number of circs. None of the big five offer a perpetual license.

Library responses:

Direct advocacy

Controlled digital lending: starting to be used. Guidelines for ILL (using DRM). Library Futures has ideas.

Legislation in Md., N.Y., and R.I. AAUP says these laws are preempted by federal copyright law. Libraries are arguing based on “reasonable terms.” Starts a conversation with publishers and may lead to federal legislation.

SimplyE: open-source software to allow libraries to federate access to multiple e-book providers. Palace Project is another one.

Publishers could offer variable licensing deals. Or even ownership as we do with print books.

Manafy: Gen X not so interested in owning media.

Blackwell: Many people are interested in owning media. Also, libraries have a preservation role.

Richard Huffine, FDIC library (formerly EPA, USGS):

News is something we all consume. Government workers consumer news voraciously. We pay for many obscure subscriptions. Sometimes it’s just not possible to get some news articles.

Agrees that publishers and journalists need to be compensated.

Sometimes copyrighted works are not available for purchase. He has printed them out and put them on the shelf. If it disappears from the internet, he still has his physical copy.

Government and corporate libraries and research publishers are part of the conversation.

Mergers, ownership by hedge funds

Open access, public access. Preprint servers. Public access is when funding agencies require open access. Creative Commons.

Also emphasized libraries’ preservation roles.

Open access copies (e.g., in institutional repositories) need to be indexed in discovery services, DPLA, etc.

Streaming content: difficult for libraries to access or preserve.

Publishers don’t understand the backlist any more. If they do make older titles available, it’s the same price as new books.

Libraries as publishers. Many university presses are coming under the university library.

Data-Driven Decisions: Citing Datasets & Transformation Toolkit

Sara Bond, JPL

Looked at citations of their datasets from PO.DAAC, oceanography datasets

Amy Stubbing, University of Westminster

Author of Data-Driven Decisions

The Future of Libraries Post-Pandemic

Kelvin Watson, Las Vegas Clark County Library

Hybrid model for programs: virtual and in-person

Plexiglass and other measures to limit touching

Digital divide:

  • Internet access
  • Hardware (tablets, hotspots)

Advocating for the library

Independent special taxing district

Digital content usage up c. 30%

Pricing models for digital content (multi-use)

Collaborating with neighboring library districts

Data Analytics for Info Pros

Frank Cervone, University of Illinois, Chicago

  • Descriptive
  • Prescriptive
  • Predictive


  • Natural language processing (e.g., chatbots)
  • Machine learning: learning what an event may signify (e.g., credit risk, fraud detection)
  • Artificial intelligence: making decisions based on the above (e.g., sending you a letter or freezing your credit)


  • Extract
  • Transfer (e.g., fix data errors)
  • Load

Data mining: processing large amounts of data to find patterns

  • Anomaly or outlier detection
  • Association rule learning
  • Clustering
  • Regression analysis
  • Classification

Machine learning: unsupervised and supervised

Black box

Ethical issues:

  • Large volume of data gathered
  • Questionable regulation
  • Unintended consequences

Privacy: de-identifying and re-identifying

Algorithm biases

Voice of the Future: Engaging & Marketing

Emily Binder, WealthVoice

Smart speakers have grown in adoption faster than any other communication medium.

When you ask for a search, it often pulls from the “featured snippets,” which often come from Wikipedia. If you want your content to be featured, use header tags, bold, boxes.

Adoption now c. 40%

Use cases: news, weather, music, asking questions

Smart speakers everywhere: cars.

Spokane Public Library put Echo Dots in libraries to educate people about bond measures

Iowa State University library used Alexa to tell about collections

Alexa flash briefings: 1-3 minute podcasts. Not that many out there. You could be number 1 in your field.

Podcast listening is still going up

If people are doing most of their internet activities with a hands-free, voice-activated device, they aren’t look at ads.

Brains are wired for sound.

Libraries Leading the New Normal & Beyond

David Lankes, UT Austin

“Libraries of all types are functioning in a time unlike any in history.”

John Perry Barlow’s declaration of independence for cyberspace (1996)

Headlines about how freedom is now limited on the internet.

Insurrection, anti-vaxers, laws against what can be taught.

The quote above isn’t true, however. These times are not unprecedented.

Not the ’60s, but 1914. Beginning of World War I, a British ship cut Germany’s undersea cables, which was the internet of its day.

Both Germany and Britain used propaganda to get the U.S. on their side. Britain censored telegrams going to the U.S.

“Propaganda” was not a bad word. It had to do with propagating faith/patriotism.

U.S. committed to “total war”: the draft, urban battlefields. Propaganda and public relations to sell it.

Espionage Act, Sedition Act. Suppressed speech seen as unpatriotic. Also, suppressed information about the 1918 flu. Hatred against unionists, Germans, etc.

Led to demonization of propaganda, isolationism. Punishment of Germany led to the Nazis.

Holocaust denial as it was happening and even now.

Misinformation can far outlast its original purpose.

Society recovered, but not on its own.

It took protestors and politicians, scholars and students to say this is not OK.

“Knowledgable people embracing diversity in a passionate drive for inclusion. Knowledgable? That’s our jam.”

We can’t wait for people to come get our information. We need to be a “distributed cadre of professionals focused on knowledge and learning.”

(I had to stop taking notes and just listen. I may have had water in my eyes.)

It got very inspirational at the end. When they post the video, I will add more quotes to this post.

Internet Librarian 2021, day 1 #InternetLibrarian

Public Health Hunger Games

Amy Affelt and Stephen Abram talked about how libraries can assist in a pandemic and other public emergencies.

Curated Intelligence: Tips and Tools

Gary Price was using Inoreader. Having issues, tested Feedly and found it better. Uses Leo, the AI assistant. Learns what you like, eliminates duplicates. Can mute phrases that you’re not interested in.

Website Watcher. Most powerful tool for following web pages. Takes RSS feeds, web URLs, and Twitter feeds.

Semantic Scholar: Can get alerts when a paper cites another one you’re interested in.

News Now: news aggregation

IFTTT: can e-mail it to yourself or others

Industry news: Cushman Wakefield, Smartbrief (for many topics)

Wayback Machine: Archive including back links. Can submit a Google sheet to archive thousands of pages at once. allows user to get RSS feeds for pages that don’t natively have them.

Search: Past, Present, and Future

Greg Notess:

Review of past 25 years of search engines. I forgot how many there were.

Google experimenting with “better” titles. Showed “Vice President Joe Biden” in a search result that came from

Google doesn’t always know when there are problems with their site.

Bing trying to make search more relevant with AI.

Current Google results give all kinds of things: news items, knowledge graph, etc.


  • Google plans to get rid of third-party cookies in Chrome in 2022. Testing alternatives. Can turn off cookies in your browser.
  • Bing and Yandex allow web sites to push info, rather than crawling them
  • Neeva, Brave: Pay for search without ads
  • Rumors of Apple developing its own search engines

Tips, Tricks, and Tools

Maydee Ojala:

Intent: search engine assumes you want to buy shoes close to you. Use footwear, the industry term

Know syntax

Use precise words

Understand scope

Advanced search: most people don’t use it (except librarians).

AI, machine learning affects search

Search is traditionally free, but search engines like Neeva are experimenting with paid search. Paywalls are more common.


  • Change word order
  • Use longtail words
  • Word order makes a difference
  • Title varies: online title can differ from print title. The Atlantic web site includes daily articles not in the print magazine or full-text databases
  • Identify fake research: tortured phrases that sound as if they’ve been translated from foreign languages
  • Comprehensive: Nothing is comprehensive and probably never was


  • Switch browsers
  • Don’t just use Google
  • Try multiple devices
  • Voice search
  • Translation, transcription
  • Beware transcription, embrace transcription


  • Old ways of evaluation were text-based, academic. Need updating.
  • Fake info on web, preprints. Moving into era of distrust.
  • What’s an authoritative source now?

New acronyms/techniques:

  • SIFT: Stop, investigate the source, find trusted coverage, trace claims, quotes, media.
  • ESCAPE (Newseum): Evidence, Source, Context, Audience, Purpose, Execution
  • SHEEP (First Draft News): Source, History, Evidence, Emotion, Pictures
  • ACT UP: Author, Currency, Truth, Unbiased, Privilege
  • Lateral reading: verify as you reading