AI and our future world #InternetLibrarian @merbroussard

Meredith Broussard, author of Artificial Unintelligence

AI is not Hollywood-style robots.  What we have is “narrow AI.”  It’s just math.

AI is a branch of computer science.  Machine learning is a sub field of AI.  It’s computational statistics, making predictions based on past data.  When people talk about being completely driven by AI with no human intuition, etc. — that isn’t happening.

Techno-chauvinism believes tech is superior to humans.  We should use the right tool for the job. When we say this, we’re saying math is better than humans.  The people who came up with this idea are white, male mathematicians from elite schools.

People embed their own biases in technology.

Female members of the American Mathematical Society: under 20%.

Word embedding: computer makes associations based on a corpus of text (e.g., occupations associated with “she” or “he”).

Funding fantasies: telling the military we’re going to make computers smarter that humans in order to get funding.  The “space elevator” idea.

Positive asymmetry: Nobody wants to be the naysayer.  Nobody wants to bring up racism, sexism, privacy.

Video of soap dispenser that doesn’t work with dark-skinned hands.  It’s a racist soap dispenser.  The creators had a blind spot.  We need more diverse teams.

Using technology is not inherently liberating.  In fact, sometimes the opposite is true.  People who don’t have computers/internet at home are disadvantaged when services go online (especially municipal services, education).

Self-driving cars are a terrible idea.  She’s read the code the tech is based on.

Video games have the same problem as the racist soap dispenser.  Facial recognition systems do not recognize people with dark skin.  They are better with men than women.  Self-driving cars are based on the same technology.

What to do?  Buy her book.

Understand AI reality.

Differentiate between AI and automation.

Assume discrimination is the default in all automated systems (e.g., AI that decides who gets welfare benefits or a mortgage).  If it’s based on past practice, past discrimination is built into the system.

Recognize ghost work.  When you flag something on Facebook, it’s mostly real humans evaluating problematic content.

Avoid tech Columbus-ing.  The study of AI is really just cybernetics, which has been around since the 1940s.  AI people need to talk to social scientists.

Read up on AI’s social aspects.

Weapons of Math Destruction

Black Software

Automated Inequality

Race After Technology

Brotopia

Behind the Screen

Algorithms of Oppression

Technically Wrong

Twitter and Tear Gas

Programmed Inequality

Talk to people:

Center for Critical Race and Digital Studies

Data for Black Lives

Black in AI

Make sure we can read today’s news on tomorrow’s computers: Turns out the Internet is not forever.  Librarian-types need to preserve content.

“The irony of writing online about digital preservation” by Meredith Broussard, Atlantic, 2015.  Legacy media organizations fired their news librarians.  They thought their CMS would save everything.  The pipes were not hooked up correctly, and there were no human beings making sure things were hooked up correctly.

Future-Proofing the News by Hansen and Paul.

Cutting-edge digital news is disappearing.  Bailiwick, her project to preserve data journalism about the 2016 election, doesn’t work any more and had to be taken down.

(Yes, she knows about the Internet Archive.  They are good at preserving static web pages, but not anything interactive or streaming.)

Archiving digital content is a human-in-the-loop process.  The fantasy is that these things can be made completely autonomous.

 

UX: Menus, Navigational Schema, & Authentication #InternetLibrarian

Aaron Bowen, Wichita State

How long should that menu be? (“Long Nav or Short Nav,” Journal of Academic Librarianship, 2018)

Testing LibGuides with students. Tested short and long versions of menu.

With the long version, students had some trouble with the tasks “Find what constitutes a scholarly article” and “How to refine a search query.”

With the short version, students had trouble with the same two tasks, plus “How to identify a research topic” and “Find information about constructing a literature review.”

Overall preference was for the long version.

Second test using software called Treejack. It strips the rest of the site out and just shows navigation menu. Confirmed results from first test.

Mohammed M. AlHamad, Abu Dhabi Polytechnic

Time for improving digital libraries UX through developing new authentication methods

Passive authentication: user doesn’t need to take any other action. Examples: IP authentication, referer link.

Active authentication: prompted to login.

If library resources are too difficult to use, users go to SciHub, Researchgate, etc.

Peer-to-peer content sharing. Springer Nature allows authors to share read-only copies of articles.

Open access articles should be clearly identified.

Some library consortiums have negotiated with publishers to pay a fixed cost for their users based on authors’ charges.

Developing a discovery layer using apps and open source #InternetLibrarian

Jarrod Wilson, Kalamazoo Public

Most web sites, the only place the catalog and web site intersect is the home page.

Open Search allowed XML calls to the ILS, which can be parsed into WordPress.  Some of the other third-party vendors (e.g., Hoopla) also allow API access.

Reduced main menu to Explore, Events, Services, Location.

Landing pages for books, audio books, e-music.

Pages with lots of detail about individual titles.  They show up in Google.  Other formats, even if they are on a different third-party service.

Collection development tool for staff.

Links for author pages, book clubs choices, genres for those not searching specific titles.

Web sites need to be accessible (anyone who takes federal money).

Branch landing page, staff picks.

Privacy: be aware of patron tracking by third-party vendors.

Able to get rid of third-party services, like Boopsie, Ektron CMS, Library Thing, etc.

Question about whether he’s concerned about too much stuff on the page.  Says no, people are used to looking at dense information-rich pages.

Nicole Carpenter, UC Irvine

Worked at two community colleges.  At one, you had to find the discovery service in a list of databases.  At the other, it was the default.

At university, had 20-year-old III interface.  Switched to Primo.

Over 50% of library resource use comes through Google.

“It’s good, and it’s getting better.”

Discovery systems are too hard to explain and teach, so the staff go back to teaching individual databases.

Went live and decided to deal with issues as they came up.

Another user group to work on back-end problems.

Different UC campuses handle search defaults, menu locations, and facets differently.

Primo is not as customizable as they lead you to believe.

Five UCs, 23 CSUs, and 110 community colleges in California going to Primo.

 

Civic Tech #InternetLibrarian

Jim Craner, Galecia Group

Apps and maps for public libraries

County property tax database: basic civic technology

Doesn’t have to be the government. Could be nonprofit or private company.

Civic Tech Field Guide

Open data
Online service delivery
Is the public using it?

Mobile device to report potholes

Citizens expect the govt. to be at most 3-4 years behind in tech

Libraries are already Civic Tech:
Online catalogs
Community tech and training
Maker spaces
Access to public info

“Third place.” De facto community centers, and you don’t have to spend money.

Computers and labs
Internet and network connectivity

Relationships with other agencies

Hackathons (e.g., map apps)

Open data catalogs (Chattanooga library did it for their city)

Historical maps (Newberry Library’s “Chicago Ancestors”)

Contra Costa Library made geo-data easier to use

Starter data:
Your own data (e.g., circ figures)
Budget
Crime
Animals (dog/cat breeds, names, etc.)
Parks and Rec

Carnegie Library Pittsburgh asked kids what they wanted to know about their city.

Code for America made application to help Californians clear their records for marijuana convictions.

“Civic Tech Brigades” in some communities.

Get diverse community stakeholders.

Remember ethics and privacy.

Local govt.: plain English instructions for things like “How do I start a business?”

Data.Census.gov

Michael Peter Edson, The Soul of a Library #internetlibrarian @mpedson

Long-time web director at Smithsonian Institution, creator of Museum of the United Nations

“Cutting the Gordian Knot, or Jibo’s Goodbye”

Marsh Mello conference on Fortnite. 10 million people attended and most of us never heard of it.

Jibo: emotional robot (like Siri only expresses feelings)

Boston Dynamics dog robot.

How do we get difficult work done in society?
How do we get millions or billions of people to work on local/global projects
Who are we and what is our future?

Climate change: 11 years to prevent irreversible damage

UN: Sustainable development goals

Story: ceramics class: half of students judged on one pot, the other half judged by weight of pots. Second group did better work

Story: MacArthur Foundation to award $100 million for solution to a global problem. Chose child welfare, but dollars were tiny compared to the problem.

Story: Asked people in Oakland, “Where do you get your culture?” People didn’t have answer. But when they asked “Who are the creative people in your community?” People lit up with answers.

Student searched “Cannery Row” on a digital library site. Disappointed none of the results were about the book, and the results were sorted by institution, not relevance. “It’s not for me,” he said.

“A problem space.”
There are many alternatives to top-down problem solving.
Even large amounts of money under central control are not enough to solve certain problems.
Common assumption about “reaching” and “serving” people are often mistaken.
There are tremendous resources among the people on the ground.

Problems often come at the beginning from mistaken premises.

“Dark Matter and Trojan Horses.”

“Delay all decisions until the last responsible moment.”

Atul Gawande: Just saying your name to somebody increases efficacy in teams.

Choose one of the Sustainable Development Goals that speaks to you. Tell your story.

Gordian Knot: Alexander the Great cut it to prove he should rule all Asia. Take the direct path. What happens more often, people take a meandering path and hope someone else will do what needs to be done. Ask: “Tell me a story about a time when that worked.”

Museums and libraries say they are activist institutions, but what are they doing about the problems they think they are solving?

After withdrawal from Paris climate accords, museum web site were doing business as usual. Teen Vogue and the Weather Channel were speaking out.

“Handoffs” in society aren’t working as they used to. Ideally, journalism should find out about wrongdoing and government should act. Everybody needs to take a little more risk.

False dichotomies: global/local, digital/physical, etc.

Gun control movement: Young people led, older people supported them.

Global is just a lot of local together.

Most institutions think of trust as something you earn, but it’s useless if you never spend it.

Dark side of tech.

Global-local
Young-old
Bottom-up
Playfully, with your neighbors, networked and at speed

Bill McKibben: “Winning slowly is the same as losing”

“If you wonder what you would’ve been doing during slavery, the Holocaust, the Civil Rights movement, you’re doing it now.”

Greta Thunberg: I don’t want you to be hopeful … I want you to act.

Postscript: Jibo is going away.

Let There Be Water: Israel’s Solution for a Water-Starved World by Seth M. Siegel: a review #water #Israel

Siegel outlines the many innovations Israel has made in the field of water and suggests that the rest of the world could benefit from its wisdom. Israel has exported its water know-how and technology to places like Africa, Iran (before the Islamic Revolution), and California.

Among Israel’s supply-side measures are:

  • Desalination of seawater
  • Desalination of brackish water in the desert
  • Intense use of wastewater (85% is reused)

Among demand-side measures:

  • Inculcation of a conservation ethos, extending even to schoolchildren and tourists
  • Drip irrigation (invented in Israel)
  • Full-cost pricing of water, even for agriculture

Siegel seems to have a bias toward free-market, capitalist solutions to water problems, but he admits that some of Israel’s innovations were developed by government programs (such as the National Water Carrier, the backbone of pipelines carrying water from north to south) and by kibbutzim, collective farms (such as Netafim, the drip irrigation pioneer).

He also admits that Israel came comparatively late to the idea of protecting its environmental water, but says that the development of the innovations listed above means that less water is taken from Lake Kinneret (the Sea of Galilee), the Jordan River, and smaller streams.

Siegel doesn’t have much to say about groundwater. He says that the brackish water extracted in the Arava desert is ancient water, which seems unsustainable in the long run. I think he also glosses over the disputes between Israelis and Palestinians over the West Bank aquifer.

However, he ends with a vision of Israelis, Palestinians, and Jordanians treating their lands as a common watershed with each providing what they can and all getting what they need. Even the Dead Sea could be saved from drying up. If all that happened, that surely would be a miracle.

Congressional Research Service reports now online from the source #CRS #LoC

tl;dr: The new site is https://crsreports.congress.gov/

The Congressional Research Service is an office within the Library of Congress that does research on current issues for members of Congress. Although it’s funded by the U.S. taxpayer, it does not work directly for us. In fact, its policy was not to make its reports public.

I first learned about CRS in library school from my reference professor, the late Dr. Terry Crowley. In those pre-Internet days, he told us that if we knew a CRS report existed on a subject, we could request it from our member of Congress.

Once the Internet came along, some public-service-minded web sites made a point of collecting as many CRS reports as they could and publishing them online. But since the reports weren’t coming from the source, you never knew if you were getting everything CRS publishes nor whether you were getting the latest versions. (CRS reports are frequently updated.)

Some members of Congress thought this was a crazy way to do things. Prodded by librarians, journalists, and other advocates for the free flow of government information, they introduced a bipartisan amendment to a budget bill requiring CRS to put its reports online. Last month, it happened. CRS reports are available at https://crsreports.congress.gov/ (Press release from Librarian of Congress Carla Hayden, Congressional Research Service Reports Now Available Online)

The site has great searching and a nice layout of results. However, it isn’t perfect. So far, it just includes reports issued in 2018. If the latest version of a report came out in 2016 or 2017, it won’t be on the site. They plan to have the full inventory online by spring 2019. Some other shortcomings of the site are detailed in this article from Roll Call, Public-Facing Congressional Research Reports Site Launches to Criticism.

So, in the meantime, you may want to check the following sites for CRS reports in addition to the official site:
 

 

 

Changing stakeholder expectations for library value #InternetLibrarian

Kim Silk, Hamilton (Ont.) Public Library
Bill Irwin, Huron University

Who does evaluation? Do you find it good or bad? One audience member was able to show that lots of different work areas used her library.

Measure to inform the strategic plan, to see changes over time, to inform our practice, to demonstrate value both quantitative (financial) and qualitative (social, educational, cultural).

Don’t confuse means and ends.

Librarians too often stop with outputs (things you can count). Also, think about outcome (qualitative results).

Public libraries often count circulation, membership, program attendance. But also consider:

Economic impact:
* value per open hour
* value per cardholder
* value per citizen

Social impact:
* student experience
* job skills
* early childhood literacy
* civic engagement
* digital learning
* economic development
* lifelong learning
* summer reading

Promotion can improve usage of parts of the collection that haven’t gotten as much use as they could.

Traditional metrics lack context. Metrics do not reflect evolution of library success. Circulation is just a means to an end.

Whatever kind of community you serve, you know which ones could use more attention. In Canadian public libraries, that’s the indigenous population, immigrants, seniors.

Engage stakeholders at every level: staff, administrators, community. Engage them about metrics.

For example, ask the children’s librarian not just to count books read in the summer reading program, but to ask the kids what they learned, what they got excited about. Maybe pre- and post-literacy evaluations at beginning and end of summer.

Question: we do surveys, but response rate is so low.

Suggestion from a corporate librarian: when you do a literature search, ask if it helped them solve their goal.

Business students learning how to use the library in their third and fourth years of school. That’s a measure. Also improved quality of work.

Market Impact: Creating Positive Outcomes & Actions #InternetLibrarian @SusanSchramm

Susan Schramm

Slides at http://conferences.infotoday.com/documents/319/0845_Schramm.pptx

Libraries are catalysts for smart communities.

Smart communities: IoT, safe, comfortable

Vision:

Streets make changes based on traffic, parking adjusted.

Firefighter knows where the people are. AR with building plans. Doctor could advise EMTs on the way to the hospital.

Open data to solve social problems

Smart parks adjust lighting

Smart airports advise you about things to do.

We used to talk about if, now we talk about how and when.

Thinking about human effects. What about people replaced by robots? What about skills we may not have enough of? What about ethics, such as algorithms that discriminate? How do we make it of, by, and for all?

Libraries can help with this.

Libraries are doing in training in skills: Maker spaces, IoT, etc. Connecting with economic development folks to reach people who want to start new businesses.

Challenges: promotion, staffing, funding. (Same for everyone)

Lesson 1: Clarify our value proposition. What problems can we focus on solving? What do we contribute? What is the call to action for our community?

Keep asking, “So what?”

Why you? Why would I get that from the library?

Sometimes it’s the how that makes you different. Libraries are safe, neutral, local.

Why now? People say, “That’s nice, I’ll have to remember that.” What’s their call to action? How can you get them to do something now?

Lesson 2: Target our audiences. Who really wants to solve this problem? (Stakeholders, not just the people for whom you are solving the problem.). What positive outcomes can be created? What communities can we partner with? Who are the influencers and early investors?

You leverage by getting your message to other groups’ members (newsletters, advertising, etc.).

Lesson 3: Help our customers buy

Lesson 4: Ask the hard questions