Competency Statement: Demonstrate proficiency in identifying, using, and evaluating current
and emerging information and communication technologies.

Introduction

Technology evolves with the help of past knowledge held in libraries, and yet libraries are perceived at being slow at adapting to new technologies. Many library institutions cannot afford to adopt and advocate every technological trend and must be just as selective with technology as they are with their collections. Identifying and evaluating technologies that will provide long-term benefits to their information environment is a competency necessary for information professionals to manage their coexistence with tools that will become ubiquitous in the future rather than being driven out by being obsolete. The following are a few emerging technologies identified by the Center for the Future of Libraries.

Artificial intelligence, Machine Learning, and Large Language Models

Artificial intelligence is an advancement toward machines that can learn and process natural language so that they “work and react more like humans” (Center for the Future of Libraries, 2024a, para. 1). Machine learning involves processing large training datasets to recognize patterns for use as predictive statistical models to generate responses. For example, Large Language Models (LLMs) is one model that processes natural language from large-scale text data. Recent demonstrations of machine learning and artificial intelligence in products such as ChatGPT have raised some concern for “routine service and clerical jobs” becoming “susceptible to automation” (Center for the Future of Libraries, 2024a, How it’s Developing section, para. 9). However, issues such as “artificial hallucination” where “a response generated by AI that contains false or misleading information presented as fact” poses concern with the ethics, quality, and reliability of AI generated content (Hallucination (artificial intelligence), 2024).

Cordell (2022) offers two opportunities to bridge the gap between library systems and machine learning. The first is a cautious nudge to integrate ML-derived data with expert-created data for “hand-assigned categories, tags, summaries, and so forth” (Cordell, 2022, p. 137). Against widespread objections to integrating machine learning, including AI hallucination, Cordell calls for libraries to assess “affordances of particular ML processes for access and discovery merit infrastructural development” (Cordell, 2022, p. 7). The second is bridging algorithmic literacy and teaching AI to the public. Querying AI, such as ChatGPT, has unlocked a new skill calledw/ “prompt engineering,” which is akin to “a reference librarian [taking] a user’s information need and translates it into an effective search query in whatever… resources offer the most relevant and authoritative information” (Bates, 2024, para. 6).

Blockchain

Blockchain is a decentralized distributed database on multiple machines that creates a data record block which is encrypted, validated by other blocks, and then added to the blockchain database. Any modification to a block must be validated by the rest of the blockchain to complete the operation. This, along with a copy of the operation ledger on every machine on the blockchain, making operations transparent and easily tracked, prevents tampering of data. Blockchain has been used to support popular cryptocurrencies, but other industries and government have investigated this technology to provide security and tracking of records, such as contracts, high-value artifacts, and personal information. However, the power consumption and emissions cost of validating large encrypted blockchains is detrimental to local power grids and the environment (Tabuchi, 2022).

As the technology works on reducing its environmental cost, blockchain technology continues to be explored, even as a solution for restrictive digital rights management (DRM), which can “restrict useful processes such as backup maintenance, accessing the content of different devices by the same owner, and so on” (Ramani S, et al, 2022, p. 1). With blockchain technology, libraries may be able to gain better access control over lending and tracking their digital assets to patrons, practicing ownership rather than having to negotiate user licenses with digital content publishers who can unexpectedly change their pricing at any given time.

Gamification

Games have an enviable position of being able to engage with millions of players, having been “credited with promoting curiosity, socialization, and the continuous processing of information” (Center of the Future of Libraries, 2024b, How it’s Developing section, para. 2). It is no wonder why education settings want to dip their feet into gamification, the application of game design elements to non-game settings. But engaging intrinsic learning is more than just “game rewards,” it’s about drawing out the game-like qualities of learning (John, 2014, para. 2). Kyle Felker (2014) reinforces this with two philosophies that allows games to do what they do best: draw in and reward players. The first values extrinsic motivators, where he introduces a hierarchy of motivators called SAPS: Status, Access, Power, and Stuff. The second philosophy relies on intrinsic motivations, or the desire to learn or explore (Felker, 2014).

Libraries make a wonderful setting for a controlled low-stakes problem-solving game environment with plenty of materials to explore using information and digital literacy on and find solutions to scavenger hunts, escape games, or other immersive experiences. As with all user-centric interactions, designing a game requires investment for user testing, research and iteration to create an engaging and transformative experience.

Evidence

Evidence #1 – INFO 247 – Python – Simple database interface

Python is a high-level programming language that is popular due to its code readability, modularity, and standard library of procedures. It “consistently ranks as one of the most popular programming languages and has gained widespread use in the machine learning community” (Python (programming language), 2024).  

In this project, I programmed an interface that ingests a formatted employee database in a text file, and gives the user options to add, remove, and view what’s in the database. The formatted text file contains a line for each employee’s last name, first name, and salary amount, delimited by an asterisk and contains no commas. In the program, the display of this information needed to be formatted to remove the asterisks, show a line number, place a comma between the last and first names, and display a dollar sign before the salary. To add or remove an employee record, the program prompts the user for first name and last name to look up.

I had to use a variety of functions to evaluate and manipulate strings, lists, and loops, which showed me the benefits of using python to manipulate and process large amounts of text data to fit a formatting standard for machine learning.

This evidence demonstrates my ability to identify and use a programming language heavily involved in an emerging technology.

Evidence #2 – INFO 246 – Text Mining – Data clustering and training

Text mining, as defined by IBM, is “the process of transforming unstructured text into a structured format to identify meaningful patterns and new insights” (IBM, 2024, para. 1).

In this text mining report, I discuss using RapidMiner to process 59 URLs within the web domains of the CDC, US National Parks, and Japan National Parks. RapidMiner is machine learning tool that uses a graphic user interface to design and implement data mining workflows that doesn’t require programming knowledge and skills (StackShare, n.d.). After processing the URLs and extracting the words from each URL into an Excel file, I used a k-means of 6 to observe how different clustering algorithms available in RapidMinor grouped the 59 websites. I found that K-means clustering yielded the best topic-related groups, which I then referenced to creating my topical category coding list. I then created two types of training datasets, one with random sampling and the other with split data sampling. In later assignments, this training data would be used to group URLs into the codified topics.

This exercise gave me a greater understanding of how text is processed for machine learning and the algorithms that can be used to categorize unlabeled data. I can see the potential of using machine learning to classify newly ingested resources for a collection.

This evidence demonstrates my ability to use and evaluate machine learning strategies to assist and augment library tasks.

Evidence #3 – Work experience – Proof of Concept Knowledge Graph

In 2024 while I was working as a game designer on Rocksmith+ (code name IBEX), I spent some time looking at how to store the complex relationships that would allow users to find songs using natural language. Luckily, I had just finished a course on Linked Data, the semantic web, and RDF, and was familiar with Protégé, so I created a music ontology using information from songs in Rocksmith+’s library. An AI company, called AllegroGraph, offered their visual knowledge graph tools for free, so I exported my ontology and imported it into AllegroGraph to create a proof of concept to show my team members how the music library search and browsing capabilities could be improved with this technology.

I created this document to explain how knowledge graphs would meet the needs of the music library, the entities and relationships used in the ontology, as well as how to get the software environment up and running to see the proof of concept in action.

This evidence demonstrates my ability to use emerging technologies to demonstrate and evaluate their feasibility and usefulness in an information environment.

Conclusion

While libraries may not have the budget to be early adopters or “on the cutting edge” of technology, I believe that keeping a finger on its pulse and knowing when a technology has stabilized enough to implement will result in long-term effectiveness in the information environments I will serve. It is helpful to have The Center for the Future of Libraries as a resource to monitor updates on technology trends as they relate to libraries, and to be able to attend sessions at the ALA’s LibLearnX conference (LibLearnX, 2022). This competency allows me to mediate technologies to patrons and provide equitable service in disseminating, educating, and supporting the community in changing and improving the ways that we live and work.

References

Bates, M. E. (2024). Librarians as prompt engineers. Computers in Libraries 44(2). https://www.infotoday.com/cilmag/mar24/Bates–Librarians-as-Prompt-Engineers.shtml

Center for the Future of Libraries (2024). Artificial Intelligence. American Library Association. https://www.ala.org/future/trends/artificialintelligence

Center for the Future of Libraries (2024). Gamification. American Library Association. https://www.ala.org/future/trends/gamification

Cordell, R. (2022). Closing the loop: Bridging machine learning (ML) research and library systems. Library Trends, 71(1).

Felker, K. (2014). Gamification in libraries: The state of the art. Reference & User Services Quarterly, 54(2).

Hallucination (artificial intelligence) (2024, October 1). In Wikipedia. https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)

Hayes, A. (2024, September 16). Blockchain Facts: What Is It, How It Works, and How It Can Be Used. Investopedia. https://www.investopedia.com/terms/b/blockchain.asp

IBM (n.d.). What is text mining? https://www.ibm.com/topics/text-mining

John, M. (2014, October 5). ‘Gamification’ Is Dead, Long Live Games for Learning. TechCrunch. https://techcrunch.com/2014/10/05/gamification-is-dead-long-live-games-for-learning/

LibLearnX (2022). Center for the Future of Libraries. American Library Association. https://www.2023.alaliblearnx.org/center-for-the-future-of-libraries

Pacific University Libraries (2024, February 16). Blockchain Technology for Library Management. Pacific University. https://www.lib.pacificu.edu/blockchain-technology-for-library-management/

Python (programming language) (2024, October 1). In Wikipedia. https://en.wikipedia.org/wiki/Python_(programming_language)

Ramani S, Sri Vishva E, Dua, L., Abrol, A., & Karuppiah, M. (2022). Blockchain for digital rights management. In SK. H. Islam, A. K. Pal, D. Samanta, & S. Bhattacharyya (Eds.), Blockchain Technology for Emerging Applications (pp. 177-205). Academic Press. https://doi.org/10.1016/B978-0-323-90193-2.00010-7

StackShare (n.d.). R vs Rapidminer: What are the differences? https://stackshare.io/stackups/r-language-vs-r-languageapidminer-language

Tabuchi, H. (2022, July 17). Cryptomining Capacity in U.S. Rivals Energy Use of Houston, Findings Show. New York Times. https://www.nytimes.com/2022/07/15/climate/cryptocurrency-bitcoin-mining-electricity.html