Successful data projects have one thing in common: they have someone who understands that technology is only as valuable as the information it serves. That person is rarely a data engineer or AI researcher - it’s the librarian, the information scientist whose profession has spent decades mastering how to make information findable, trustworthy, and useful. Yet most digital transformation initiatives never bring them to the table. We hire for technical expertise while ignoring the one professional trained to turn information chaos into human understanding. Libarians are the real Knowledge Modelers.

The Librarians Who Taught Me to See Differently

I learned this lesson early. In the late 1990s, fresh out of the Air Force, I took a job as a PC tech working across several school districts in the Eastern Upper Peninsula of Michigan. I thought my job was simple: fix computers, install software, keep the network running. I was a technician focused on the machinery.

But it was the school librarians - who were rapidly becoming “media specialists” - who showed me that technology work was really information work. While I was crawling under desks running ethernet cables, they were asking the questions that actually mattered: What will teachers be able to do with this network that they couldn’t do before? How will students find the resources they need? What happens when the technology breaks - do they know where to look?

These librarians were leading the digital transformation of not just their school districts, but entire communities. The libraries became hubs - places where seniors learned to use email, where students researched beyond the encyclopedia, where teachers discovered new ways to connect information to learning. They understood something I didn’t yet grasp: the technology was never the point. Access to curated, contextualized, trustworthy information was the point.

Over the next few years, I watched librarians across multiple districts evolve from “media specialists” into de facto technology coordinators. They didn’t just manage the card catalog; they architected entire information ecosystems. They trained teachers. They evaluated software. They designed search systems. They asked the hard questions about equity: Who can’t access this? What barriers exist? How do we serve everyone?

Those librarians became some of my most important mentors, shaping how I approach information technology and data science today. They taught me that before you build a system, you need to understand what information it should serve and why. That lesson has followed me from school networks to rural broadband planning to AI-powered knowledge graphs.

The work hasn’t changed - we’re still trying to connect people to the information they need. But somewhere along the way, we forgot to bring the librarians with us.

The Problem We Keep Ignoring

Imagine this scenario: You’re working on broadband infrastructure planning in Michigan’s Eastern Upper Peninsula, sitting in a room with local and state officials, ISPs, and federal grant administrators trying to determine which locations qualify as “unserved.” You have the data. You have the maps. You have brilliant engineers. But you can’t answer a seemingly simple question: “Why do we have three different definitions of ‘household’ across these datasets, and which one should we actually use?”

The FCC data says one thing. The NTIA says another. The census says another. The housing authority has a third definition that reflects the actual lived reality of multi-generational homes in the community. The data engineer points to the API documentation. The GIS specialist shrugs. The program manager says “let’s just go with the federal standard.”

But imagine if a librarian - someone embedded in the community - was at that table. They would ask the questions that matter: What did each agency mean by ‘household’ when they collected this data? Which definition aligns with our equity goals? What cultural assumptions are baked into each measurement? Which data actually reflects how people live here?

That conversation would change everything. You wouldn’t just be merging datasets - you’d be reconciling fundamentally different ways of understanding community. The librarian wouldn’t just help you pick a field; they’d help you understand what you’re actually trying to measure and why it matters.

This hypothetical captures the crisis hiding in plain sight across every digital transformation initiative I’ve worked on: we are drowning in data but starving for wisdom.

In my work at the intersection of information and technology - from planning smart community infrastructure in rural Michigan to designing knowledge graphs for digital literacy and skills - I often return to a simple mantra: LearnIT | BuildIT | TeachIT. We are very good at the “building” part. We can lay fiber, spin up servers, and train models. But as our “digital laboratories” grow more complex, we face a challenge that code alone cannot solve: curation.

This is where the modern librarian - the information scientist - becomes not just helpful, but mission-critical. Not as a support role. Not as a documentarian. But as the architect who ensures our digital systems serve human understanding.

1. From Data Lakes to Knowledge Graphs

My current focus on Knowledge Graphs and AI has reinforced a truth that library scientists have known for decades: context is everything. You can feed a Large Language Model a terabyte of text, but without ontology, taxonomy, and semantic structure, you are merely building a stochastic parrot.

An information scientist brings the rigor of metadata management and classification theory to our chaotic data pipelines. They don’t just store data; they structure it. When we build systems to personalize learning for students, we need more than just raw code - we need a schema that understands the relationships between concepts. We need the “card catalog” of the 21st century, ensuring that the right information gets to the right learner at the right time.

2. The Ethics of “Learning in Public”

I treat my work as a practice of “learning in public,” using a status system to move ideas from Research to Draft to Review to Published (which is an experiment in itself). But in an era of misinformation and algorithmic bias, this transparency comes with a heavy responsibility.

Librarians are the original guardians of information integrity. In my work with the Digital Equity Learning Project, the stakes are high. If the data guiding our broadband policy is biased or unverifiable, real communities get left behind. An information science professional serves as the ethical check on our technical ambition, ensuring our sources are cited, our data is traceable, and our algorithms respect user privacy. They transform “open source” from a code repository into a trustworthy archive of public knowledge.

3. Accessibility as a Design Principle

Technologists often confuse “availability” with “accessibility.” We might build a high-speed network (availability), but if the interface is unusable or the information is unfindable, we haven’t achieved digital equity.

Librarians are trained in human-information interaction. They understand that a system is only as good as a user’s ability to navigate it. Whether I am designing a passive optical LAN or a data visualization for rural school districts, the library science perspective asks the difficult questions: Who is this for? How will they find it? What barriers stand in their way?

4. What Happens When We Build Without Them

The cost of ignoring information science is not theoretical. I’ve seen it firsthand:

  • The Knowledge Graph That Ate Itself: A team spent 18 months building an ontology for educational resources. Without a librarian’s understanding of controlled vocabularies, they created 47 different terms for “mathematics” across their schema. The graph became unusable.

  • The Open Data That Nobody Could Use: A city released 15 years of transportation data as “open source.” No data dictionary. No provenance. No context. It sat unused because nobody could figure out what the column headers actually meant.

  • The AI That Couldn’t Explain Itself: A model made high-stakes decisions about digital equity funding, but the team couldn’t trace back which sources informed which recommendations. When auditors asked for citations, the engineers had no answer. The librarian would have built that traceability from day one.

These aren’t edge cases. This is what happens when we treat information as a byproduct of technology rather than the purpose of it.

The “Info” in Information Technology

For too long, IT has obsessed over the “T” - the technology. We fixate on the pipes, the fibers, and the GPUs. But the “I” - the information itself - is what actually matters to the people we serve.

In my journey as a data scientist and advocate, I have realized that I don’t just need better engineers; I need better architects of meaning. I need the librarian who understands that while code runs the machine, information science is what makes it human.

How to Actually Work With Information Scientists

If you’re leading a digital transformation initiative and don’t have information science expertise at the table, here’s what to do:

  1. Bring them in at the design phase, not after the data warehouse is built. Let them shape the ontology, not just document it.

  2. Give them decision-making authority over taxonomy, metadata standards, and data governance. This isn’t clerical work - it’s architecture.

  3. Pair them with your data scientists. The best data science happens when statistical rigor meets information literacy. One knows how to build the model; the other knows what it should mean.

  4. Listen when they ask “why?” When a librarian questions your data structure, they’re not being pedantic - they’re protecting you from building on a foundation of assumptions.

  5. Pay them like the professionals they are. If your data engineer makes $120K and your librarian makes $50K, you’re signaling that the meaning of your data is worth less than the storage of it.

The next time you kick off a data initiative, ask yourself: Who on this team can explain what this data means? Who will ensure it can be found, understood, and trusted five years from now?

If the answer is “nobody,” you don’t have a technology problem. You have an information problem. And the librarian is the only one who can solve it.