Podcast: Play in new window | Download
Subscribe: Apple Podcasts | Spotify | Amazon Music | Android | Youtube Music | RSS
After 20-plus years of industry analysis, Alan Morrison has developed a keen sense for how knowledge graphs can help enterprises.
Even though he has focused on advanced tech and emerging IT practices and is deeply immersed and invested in current tech developments, much of his advice for enterprises looking to develop their data maturity involves pragmatic baby steps and basic mindset shifts.
We talked about:
- his work in the consulting world and his organizing work around the knowledge graph community to improve awareness of the technology
- the need to find “foxes instead of the hedgehogs” in enterprises when you’re trying to promote adoption of new tech
- the relationships between different AI tech, like LLMs and knowledge graphs, and the common connection they share: data
- the importance of having mature data practices in any enterprise
- how even simple metadata practices in common tools like spreadsheets can support better enterprise data practices
- how sidestepping the formal org chart and forming guerrilla teams can advance data practice
- the benefits of starting small in any knowledge graph project
- how representing organization knowledge at a high level in a knowledge graph can help solve big enterprise problems
- how a knowledge graph gives you a multidimensional Tinker Toys set to model and understand your org’s data
- the benefits of moving from tabular thinking to graph thinking
- his frustration with the current framing of AI as being solely about machine learning
- his observation that practices across any org – content, knowledge management, data management, business people – could benefit from long-standing standards and proven technologies (that might not be as sexy and topical as LLMs)
Alan’s bio
Alan Morrison is a longtime analyst, writer, advisor and podcaster on advanced data technologies and emerging IT. For 20 years at PwC’s R&D and innovation think tanks, Alan identified emerging technologies on the cusp of adoption, assessed their business impacts, and advised PwC’s clients on innovation strategy.
Before PwC, he was a semiconductor industry market analyst and forecaster, a retail site location analyst, and a US Navy intelligence analyst, Russian linguist and aircrewman.
For the last five years, Alan has been a contributor on knowledge graph and related topics for Data Science Central. His writings over the years have covered dozens of different technologies.
Connect with Alan online
Video
Here’s the video version of our conversation:
Podcast intro transcript
This is the Knowledge Graph Insights podcast, episode number 5. You might think that the lofty perch of multiple decades in industry-analyst roles would inspire grand visions of tech transformation with leading-edge technology. Quite the opposite in the case Alan Morrison. He shows how enterprises can advance their data maturity by cultivating basic graph thinking in their organizations and by taking small, pragmatic steps like adopting established standards for interoperability or simply adding metadata to a spreadsheet.
Interview transcript
Larry:
Hi, everyone. Welcome to episode number five of the Knowledge Graph Insights podcast. I am really happy today to welcome to the show Alan Morrison. Currently, he’s a contributor at Data Science Central, a well-known publication in the field. He’s a freelancer and consultant around knowledge graphs and a lot of other areas as well. His background, he comes out of the consulting world. Most recently before his current role as a freelancer and consultant, he worked for many years at PriceWaterhouseCooper, the big consultancy as a senior research fellow. So welcome, Alan. Tell the folks a little bit more about what you’re up to these days.
Alan:
Hey, Larry. Great to talk with you and folks should know that you and I have some history together as a part of the Data Worthy Collective, which is like an informal meetup group, collaborative thinking going on every week, and it’s been great to know you over the years. What I’m doing currently is trying to help the knowledge graph community, in particular, gain more visibility, gain more traction in the enterprise, and it’s a long haul. Enterprises are slow to change. It’s like turning an oil tanker on a dime. It’s very, very hard kind of thing to do, and there’s so much legacy involved. And so when I was at PWC, we worked with a lot of large companies and you’d look for pockets of innovation and you’d look for the foxes instead of the hedgehogs, because the foxes were the ones that were curious about doing things different ways. The hedgehogs were the ones who were expert in doing things in one way.
Alan:
So we had this kind of guerrilla approach to innovation, and we also worked with the centralized innovation group inside the firm to try to help the firm itself modernize. And so in my last five years at the firm, I was plugged into the AI efforts that were emerging because machine learning was becoming much more feasible. And so I’ve taken this knowledge that I have of AI and the semantic web so-called, an old term, but it still has utility, together, and I’m just trying to help enterprises see the advantages of these things and adopt them to the extent they can be.
Larry:
You just said how hard, notoriously difficult it is to get enterprises to think and act differently, but they’re all jumping all over AI like it’s the best thing since sliced bread. But there’s a lot of opportunities there, it seems like, to not ride the coattails, but to enter the conversation around these new technologies. And there’s a lot of interplay between generative AI and machine learning and LLMs and all that world and the knowledge graph world. Can you kind of stitch those worlds together for us a little bit?
Alan:
Yeah. I think it’s good to do that. It’s good to step back and say, “What are we trying to do here? How are we trying to do it?” We’ve got some piece parts that are talked about in the media at infinitum. Generative AI is just all the time in the conversation because it’s a powerful interface technology, as it stands, and some enterprise providers are using GAI as basically a front end, and then they will connect their own backend. And so I think you have to think about generative AI and other kinds of AI in the context of this bigger picture, and it’s all driven by data. And when we say data, we don’t just mean binary bits, I think we mean ideally contextualized information, knowledge, getting wisdom and decision-making capability to the right point where it’s actionable at the right time for the right purpose.
Alan:
And so it’s a distribution problem of knowledge, essentially the right kinds of knowledge. And so you really have to think about data, and this is where I get passionate about it, because the information, the heart of it is in this contextualized environment that should be being built, and it should be an organic kind of effort that involves both humans and machines. And I’m a woodworker, so I think about machine learning as a kind of a table saw. And so the knowledge is the wood that you’re working with, it’s an organic thing. And so you’re using all this different kinds of tooling. I got a lot of different kinds of tooling in my workshop, and I’m not a great woodworker, but I think that the source of the wood is really important, what kind of wood you’re working with. And we could do all sorts of things if we had the right resources inside of enterprises.
Alan:
Enterprises are really starved for good data. They just are not in the habit of collecting it very well. I was in intelligence in the Navy when I started, just collecting voice traffic and analyzing it. And it was so systematic about how the data collection happened. There was this whole data lifecycle environment that we were a part of, and everybody was managing according to the needs of that. And I just think that that was a very effective way that enterprises could take advantage of to really collect what they need to and just understand if you’re going to digitize things, you have to have this continual process of collecting and analyzing and managing this information. And it has to be organically constructed so that it’s scalable and it does what you need it to do. So that’s basically where I’ve been focused over the past years.
Larry:
Yeah. The way you described that is so evocative of, I love the word working analogy, but also your military experience. I think any enterprise would argue that they would claim to attribute the same importance or similar level of importance as naval intelligence data about whatever you were researching. And yet, they have these sloppy data practices, or if not sloppy, at least not thought-through and sort of suboptimal. Can you talk about two things? One, how could they be better at that data hygiene and that data practice that you just described that was so well entrenched in your Navy days? And then in particular, how knowledge graphs and the whole world of semantic tech can help you do more with that data, and is it a prerequisite to have that good data hygiene before you can do the cool stuff with knowledge graphs?
Alan:
Let me start with the last question first. It is a prerequisite to have a certain amount of data maturity. When I was at PWC, we had a data maturity curve, and it was frustrating to see that most of our audit clients were not terribly high on that maturity curve. I think the tooling gets in the way. There’s so much siloing that has gone on over the decades, and so many folks, including me, are in the habit of just using certain tools. And so the learning curve for learning something new is a bit steep. And so what’s happened is that we’ve just proliferated these data silos that have limited utility and a short lifetime when you could be just contributing to a much larger ecosystem that’s shared and reusable. And so I think that there’s just a mentality that goes along with the tooling that needs to be dispensed with.
Alan:
So you could start up, say, a guerrilla team that just tries to do these domain description kinds of exercises. You’ve seen pharmaceutical companies do this for years and years because they have to do drug discovery at the molecular level and they have to track things at the molecular level, so it’s very precise and detailed what they’re doing. And they’re still using spreadsheets, but they’re describing their findings in the spreadsheets, and then they’re making the spreadsheets part of this bigger knowledge graph mix, if they’re advanced.
Alan:
And so what you’re seeing is the ability to do more metadata, let’s say, semantic metadata, simple stuff, not complicated stuff, and just put enough of it in a spreadsheet so the spreadsheet is shareable and reusable in a sense. You’re seeing this in financial services, too, where we’ve got the OMG, the Object Management Group, is working on a standard for smart spreadsheets. So we’re trying to desilo it to the extent we can. We’re taking it a step at a time.
Larry:
That’s interesting. And when you talk about silos and guerrilla teams, and there’s this truism in consulting about just start small and just start someplace and anywhere. I’m curious about these guerrilla teams that you would assemble back in your consulting days. I hope that a lot of my listeners are going to be people who are in organizations that are like, “Oh, come on. We got to do something with this,” and they’re going to be chomping at the bit. They might have big ambitions. So I guess first of all, and I am as guilty of this as anyone, first scaling back your expectations and your ambitions, but also with the awareness that if you crush it at a small scale, that you might get bigger adoption. So I guess my question is what do those guerrilla teams look like and how do they work?
Alan:
I sort of backed into this years ago when we started writing about the semantic web so-called back in 2009. We were publishing a quarterly called The Technology Forecast, and we were looking at business intelligence. What’s the stumbling block in business intelligence? And of course, it’s having enough of the right data to make good decisions. So you have to bring in all these different sources, these heterogeneous sources, not only structured data, but less structured information. And so that was the starting point for thinking about what organizations should do with their data.
Alan:
But within PWC, once I had researched the topic, I got so interested in how this could be really a big thing and could really help companies if they were to bring together all this heterogeneous information in a more integrated and interoperable fashion. So I started working with client companies and internally at PWC with innovators who were as passionate as I was about organizational change and bringing these organizations up to speed in certain ways.
Alan:
So we’re really sort of working against the conventional org chart, the hierarchical org chart. One of the things we knew to do was to sort of build our own informal organizations. So we had the flexibility in a consultancy to sort of be tribal in how we organized, and we built our own tribes, and partners who were powerful partners would be the tribal leaders. And so you’d look for partners who had budget, who could fund innovation projects.
Alan:
And so I think the same dynamic happens in a command and control environment, like the military, for example. I’ve seen champions emerge, for example, in the Department of Homeland Security, and people who have a Border Patrol background working to advance the cause of adopting knowledge graphs for border security and detecting and intercepting drones, for example, on the border when they’re trying to smuggle drugs or guns or something. Same principle applies. It’s just harder to do inside a command and control environment than it is in a loosey-goosey thing, like a partnership, for example.
Larry:
Yeah, I love that guerrilla approach. Kind of what you’re getting at, too, is about the power and the benefits of a knowledge graph. You were talking about the heterogeneous data, the structure, unstructured, just sort this mishmash of stuff. Can you talk a little bit about how a knowledge graph helps pull that, or maybe not pull it together, but pull information about those kinds of bodies of data into a way that is more powerful and useful to the organization?
Alan:
It’s a great question. The experiences that I’ve seen have mostly been failures when it comes to these kinds of efforts over the years. They’ve been useful failures, and I’ve had good friends who’ve been involved in these failures, but the situation has been that we’re trying to boil the ocean when it comes to a model of the data, a unitary model of the data that brings all the data together so that you’ve got a model of the business essentially in an enterprise. And this is something that Dave McComb at Semantic Arts preaches all the time. You want to rationalize that model. So I saw a model at IBM back in the 2010s when they were working with ontologies. They were trying to get all their software together, all their software development efforts together with a knowledge graph at that point, and they had over 1,000, 2,000 concepts or something. Just a monster thing.
Alan:
FIBO was another effort, Financial Industry Business Ontology. Just a massive effort of these big banks to try to model the environment that they needed to use for reporting purposes, for example. It’s much better to have a few hundred concepts at a maximum and then focus on domain specific efforts that will articulate the domain so that you start to get a sense of these contexts that I was talking about and how the contexts fit together. So the knowledge graph is essentially a relationship-rich approach to connecting these things and studying how they interact.
Alan:
And the thing about the way AI is currently done, by comparison, AI currently with machine learning, it’s sort of like a recipe in a kitchen. You put some of this ingredients and some of those ingredients in together and you mix it all together in a mixer. This is not really helpful for disambiguation, comparing apples to apples, certain kinds of apples to apples or certain kinds of molecules to other kinds of molecules. Knowledge graphs have to be articulated in the relationships in the entity relationship models are what makes the articulation possible.
Alan:
I’m married to Lee, so the fact that I’m married to Lee leads to all sorts of interesting things about our history together, and that history together is a context. My family is a context. And so similarly with organizational change, you want to be able to model the organization, the people inside the organization. You want to empower them. And so it’s good, it’d be great if you could have the organizational change people get together with the knowledge graph people, build a knowledge graph about the organization, and sort of empower the innovation. Do that kind of thing we were trying to do back in the day with our innovation groups.
Larry:
That’s interesting. You’ve mentioned you’ve written fairly recently on LinkedIn. I thought it was a success story about using a knowledge graph is … I don’t know. I think in a lot of my consulting and contracting work, I think of all my deliverables not as final deliverables, but as conversation starters or conversation continuers, that it’s never done and you never figured it all out. And so when I read that post where you were talking about knowledge graphs, that’s how I was kind of picturing it like, “Hey, let’s gather around this knowledge graph and see if it helps us understand the org better.” Was I reading that right?
Alan:
Oh, absolutely. You have a whiteboarding dynamic ideally where you’re talking about the entities and the relationships. So you’re representing, you’re basically representing at a high level the organization and the problems you’re trying to solve together. How can you best utilize the resources of the organization to solve certain problems? People do this all the time on whiteboards. They’re always drawing diagrams on whiteboards. You always see mind-mapping going on. But the knowledge graph is just an extra step that articulates the digitization of it so that machines can help humans do more with it. You can actualize the model that you’ve come up with on a whiteboard in a knowledge graph environment. That’s where we should be going with this technology is kind of a Miro kind of thing that people can work on together and then activating it. That way, the enterprise architects would be much more empowered to do what they’re supposed to be doing.
Larry:
Yeah, boy, you’re preaching to the choir. I’m working with a friend now on trying to get the precious data out of Miro boards into something more manageable and useful. But just the way you talk about that, I love that description of that these are activities we’re doing anyway. And I think one of the appeals of graph technology is the nodes and edges, just nodes and connections between them, which is very much like the things you sketch out on a whiteboard in a workshoppy kind of session. The power of the graph, and you just alluded, you just mentioned this as well, is that it’s articulatable, those entities and their relationships are articulatable in a way that both humans and machines can understand, which I think is kind of the fundamental power of the knowledge graph. Is that correct?
Alan:
Yeah. Certainly, that’s a big part of it. Being a woodworker by hobby, I think of it as sort of Tinker Toys, and I don’t know if many people know what Tinker Toys are, but you’ve got a wooden hub with holes in it, and then you’ve got sticks that you used to build something, and you can build any kind of three-dimensional object with tinker toys. And in a knowledge graph environment, the only difference is that you’ve got the potential for all sorts of different dimensions, and humans can only comprehend a few dimensions at once, but machines can go far further. But you can sort of design the interface so that humans have what they need and machines have what they need, and then there’s a coming together of those understandings in the enriched information and the data and the metadata.
Larry:
Yeah, and that interaction. I want to swing back a little bit to the GenAI stuff and LLMs, because you mentioned earlier that that’s sort of the interface for a lot of this now, and there’s emerging architectures, it’s like you’re trying to get the best out of each of those kinds of technology. People love the conversational interface and conversing with their data aspect of an LLM, but all those benefits, you’ve just been talking about knowledge representation with a knowledge graph. It seems like there’s been just the last week or so Microsoft announced their graph RAG scheme, and there’s many other RAG kind of things out there. What do you think is going to be the most productive way or the kinds of architectures that’ll emerge to optimize the benefits of these two sort of complementary technologies?
Alan:
Yeah, I think there needs to be a fuller picture of all the things that can contribute the most to this kind of environment. And when you’re focused on the knowledge graph space, you learn a lot about different kinds of databases. And so the database people in this space are really super smart and they’re doing some really interesting things, and that stuff should not be neglected. I think that that stuff is just as valuable as what’s going on in generative AI and neural networks.
Alan:
And I think that one of the frustrating things about web development and data science is that there’s not enough attention paid to all the techniques that are still incredibly important today that have been around for decades, and database technology is one of the primary ones that I would point to. And so the difference in databases now versus 20 years ago, relational databases haven’t scaled as well as graph databases do, and the graph databases help you change the mentality of the people who are working on the things.
Alan:
I’m sort of rambling here, but let me just finish the point. In the knowledge graph community, there are people who think in graphs. Not everybody thinks in graphs, and tables are simpler, a lot of people can utilize tables, everybody uses tables. But the fact that you’ve got this graph mentality leads to connectability and scaling and ecosystem development, and where we should be going with this, ultimately, Larry, is not just feeding some data into a model and trying to evolve the model. It’s really building systems where we can share whatever we want to in a digital form that’s consumable by machines and humans at the same time.
Alan:
So we need more than just neural networks to be able to do that. And I think that a lot of these technologies that have been around for 50 or 60 years are complementary to what’s happening in machine learning. And what’s frustrating to me is that a lot of people just think AI is just machine learning, and intelligence is so much more than just machine learning.
Larry:
Absolutely. Knowledge graphs, knowledge representation is a form of AI, and expert systems and robotics and vision, and there’s all kinds of other stuff. Now, I want to do a whole other episode just about the database technology that underlies this, but today, we’re running short on time and I like to keep these around a half hour. But before we wrap up, is there anything last, anything that you want to revisit from the conversation or just want to make sure we share before we wrap up?
Alan:
Yeah. I think that in general, there’s so much opportunity out there. I don’t know where to get started, and sometimes I don’t know who to talk to. I’ve been looking at systems and how they’re coming together, and so I’m talking to a content person when I’m talking to Larry, and I talk to knowledge management people, and I talk to data management people, and then I talk to business people, and all of these people could be using the same system to manage information and make it shareable. All of these people could be doing things the same way. We have the standards, they’ve been solidified.
Alan:
We are really blessed with a community in terms of semantics that has the vision to bring this system together, and what we need is just more oomph to bring the organization into the picture more somehow and make them understand. You can just have one department. You don’t need three departments to manage all this stuff. And why don’t you all work together the same way? I wish we could have a meta organization capability that we could apply to all these companies that have these issues.
Larry:
I love that vision of a meta organization, like the, not overarching, but just the connect-ey part of it, which is what graphs are so good at.
Alan:
Yeah. Collaboration is something that, that’s another conversation itself, isn’t it?
Larry:
Yeah. Well, this is only episode-
Alan:
That’s scale of collaboration.
Larry:
I was going to say, this is only episode five, and I definitely want to have you back on for several more conversations, because there’s five rabbit holes we didn’t go down today that I would love to go down. Well, thanks so much. Oh, hey, one very last thing, Alan. If folks want to connect with you online, what’s the best way to find you?
Alan:
I tend to look at LinkedIn every day. I hope I’m findable on LinkedIn, A-L-A-N.
Larry:
I’ll put it in the show notes as well. You’re one of the more prominent Alan Morrison’s there, I’ll observe.
Alan:
Well, me and an organist and a scientist are the ones, so I think you could figure out which one’s which.
Larry:
Yeah, I’m duking it out with a brain scientist for the Swanson crown. I don’t know. I don’t like my odds. So anyhow, thanks so much, Alan. This is a blast.
Alan:
Sure. I’m happy to do it.
