Ren Pope is a Principal at Info-Do, which offers data, information, and knowledge architecture consulting. He’s worked in these domains for the last 25 years or so, and has also taught colleagues how to do these things through workshops and presentations. In this episode, we focus on why organizations need to know about and develop ontologies.
- Ren Pope on LinkedIn
- Ren Pope on SlideShare
- The Information Architecture Conference (previously The Information Architecture Summit)
- Jessica DuVerneay on LinkedIn
- Hyponymy and hypernymy
- Sentence diagram
- Card sorting
- Mind map
Some show notes may include Amazon affiliate links. I get a small commission for purchases made through these links.
Read the full transcript
Jorge: So, Ren, welcome to the show.
Ren: Thank you very much! Glad to be here. Thank you.
Jorge: Well, I'm very excited to have you on. For folks who might not know you, can you please tell us about yourself?
Ren: Sure. I'm an information architect based out of Washington, DC. I've been working with information, data, and knowledge for about 25 years, mainly within the US government, but a little bit in research and development. And for the past year, I've been branching out into the commercial sector, doing some work within healthcare and publishing and other content driven endeavors. I've had a lifelong, I guess, passion for information and data. When I was younger, I didn't know it, but my mother was a healthcare statistician and data scientist. All I knew is that she did a lot with computers and crunching stuff. And from there I guess I got the data bug from her, and she helped me build my first database when I was about 13 or 14, for a high school project. And ever since then I've tried to organize things, tried to make good information for myself and present it to other people.
Jorge: That seems very apt to the situation that we're currently finding ourselves in, especially healthcare statistics.
Ren: Definitely. I was fortunate enough she let me do an internship at her company, which was, one of the precursors to Cigna Healthcare, which is a large healthcare provider. And I worked in the database shop, building database structures and then documenting them as a technical writer. And, it's interesting healthcare data and information. And the better information that you have, the better decisions that people can make.
Jorge: You use the phrase " good information" before, and I'm wondering what constitutes good information for you?
Ren: Well, I think good information starts with good data and data is the ingredients that make up information. On its own, you can't use data. It has to have context. And that's what information is. And this is my personal belief and philosophy. So, to have good information, not only do you have to have that good data, but you also have to have it put into the right context for that situation. And that's where you get into semantics and a lot of other things, but good information definitely boils down to the data, the way that it's structured in the way that it's presented.
Jorge: I'm writing down notes as you're speaking, and I wrote down the formula: data plus context equals information. Is that a fair summary of that idea?
Ren: Yes. And then that's surrounded by, let's just call it meta, the metadata, which points you to how to take that data and make it into information.
Jorge: Some folks listening might not be familiar with metadata. How would you describe that?
Ren: Sure. So, metadata is data that describes other data. So, Ren Pope is my name. That might be data. If you didn't know Ren Pope was a name, the actual word name puts into context, Ren Pope, as a name. So, name would be the metadata for Ren Pope.
Jorge: When I reached out to you about being on the show and I asked you what we should focus on, you used a phrase that piqued my interest; you said, "ontological thinking." And I'm wondering if you could elaborate on that.
Ren: Sure. Well, in 2016, at the Information Architecture Summit, I did a talk on ontology, which is, a branch of information science that information architects use to help understand and describe information. And I just gave a basic talk about this at the Summit and in the audience was Jessica DuVerneay who took my talk and ran with it and used it to help her in her information architecture practice. The next Summit, she came to me very excited, kind of gave me an insight onto what she was doing and we decided to work together to come up with a workshop and she came up with the term ontological thinking and that was a main part of the workshop. And how to practically apply ontology in information architecture and, I guess, a broader term, into business thinking.
Jorge: So, ontology is one of those terms that I find when I use it with folks, I often have to explain it. It's a term that I think many folks are intimidated by. And I'm wondering, how do you introduce this subject to people, especially people who might not be familiar with information architecture.
Ren: Sure. Well, the funny thing is, is that I introduce it in a way that's kind of humorous because nine times out of ten -- especially in the medical field -- when I say that I do ontology work or that I'm an ontologist, they say, "You work with cancer?" And, I say, "No, no, no, no, that's oncology with a C. This is ontology with an N and a T." And, it's really the study of things and how they relate to other things. And ontology comes out of the schools of philosophy from Greece. And back then, ontology was really to determine if something was real or not. And through the centuries, ontology has grown into, not only is it real, but is it something that's concrete or something that's actually physical? Or is it something that's more conceptual, or abstract. And then the, how things relate to each other, has become a much more important part of ontology than it was back in classical times. And now we have applied ontology where it's not just a philosophy, but actually a scientific pursuit where information scientists can use these philosophical thoughts to help organize information, understand information, and explain information.
Jorge: You talked about applying ontology, particularly in business contexts, and I'm wondering if you could give an example of how an ontology might be useful in a business context.
Ren: Sure. Well, another important thing is, is that in this modern day of, of ontology, you said something very important, you said, "an ontology." So, ontologies now have become a noun, where you can have, the applied efforts of ontology encapsulated into an ontology. And normally that goes against a domain. So, it could be medicine, it could be business, sports, any domain that you can think of. And to use the example of medicine, since medicine is very complex, it's even complex for healthcare providers such as doctors. You have many different types of medicines that address many different types of illnesses, and to make sense out of all this madness, ontologies can help describe the relationships between medicines, doctors, patients, conditions, symptoms and all of the aspects within healthcare.
Jorge: I'm going to trot out another kind of big word that information architects use all the time, "taxonomy," and contrast that with ontology. So, when you're saying the relationships between things like medicine and doctors, I know that a taxonomy has to do with the types of medicines, for example, like an official list of the medicines that would be available within that context, whether it's a particular hospital system or whatever. Am I right in understanding that the ontology would be kind of the higher-level relationship between the categories of things?
Ren: Yes. So, there's different schools of thought within our community as what the relationship between ontologies and taxonomies are. And the really funny thing too, normally I follow up with trying to explain ontologies after I talk them off the ledge that I deal with cancer, that I deal with taxonomies. And they say, Oh, so you stuffed animals? No, that's taxidermy. So, in my view, taxonomies are very specific applications of an ontology. So, an onlology looks at how things are related, whether they are a part of something, whether they are related closely to something, whether they are not something. So, an apple is not an orange. And a taxonomy is focused on specifically types. A type of. So, the most famous taxonomy is the biological taxonomy. So, a cat is a type of feline. A feline is a type of mammal, and a mammal was a type of animal. And it's that very specific kind of line from animal to cat. An ontology goes a little bit broader. A paw is a part of a cat and a cat, eats meat. That kind of breaks the rules of a taxonomy, but it is accepted within an ontology.
Jorge: The way I'm hearing that in that particular example is that the taxonomy is a structure and I'm visualizing like an outline. You mentioned the taxonomy of the cat as belonging to a category called feline, which is in itself a part of another broader category called animal. And that I visualize kind of as an outline, and you would expect that there would be other branches or other kind of outline items that would be the peers of the cat. And they would be things like, in the case of felines, like tigers or panthers or what have you. But the point is like, they're all kind of about the same type, whereas the ontology is describing the relationship between things that are related, but they're not the same thing?
Ren: Not quite. So, when I build an ontology for a client, taxonomies will be a part of that ontology. It is one dimension, if you will, it's the type of dimension. And then, along with that, think of it as almost like an ecosystem. You're describing that ecosystem of cats... Let's just say, the ontology of cats. You want to know the different types of cats. You want to know the anatomy of the cat, so the parts, different parts of cats. And then you want to know kind of, so how does a cat live? What does a cat do, to get along, to be a cat? And that could be, what it eats, how it interacts with other animals, where it sleeps. So, the ontology is much broader, and the taxonomy is just a very specific part of it. So, there's some very, I guess scientific words that, when you really dive in into ontology to explain those different things. So, taxonomies deal with what's called hypernyms. And then a hypernym is just getting to something very specific, the very specific name. So that's what hypernyms means. Hyper, very specific. Nym, name. And then you have meronyms, which are the parts of mero, meaning part, and then name. So, the part name of something. So, like a paw, a part of the arm or the leg, and then the leg is part of the body. So, you have these, what I call nyms. The most popular "nym," that everybody's used to, are synonyms. So, the like name. So, a tree is a synonym to a plant, although you could kind of argue that it's also a type of, but in certain contexts, they are synonyms. So, you really, you're dealing with the different name issues surrounding a subject.
Jorge: You know, I'm hearing you describe these things and feeling like these are concepts that we are all familiar with in our day-to-day lives, and you're giving us a language to talk about them. You know, the idea that the paw is a part of the cat is something that I think comes intuitively to a lot of us. But people might not be aware that there are formal ways of describing those things, those types of relationships.
Ren: Yes, and that's the thing. So much of this is subconscious that on a day to day basis, you don't even have to think about that. But when you're making very complex systems... So let's say if it's an e-commerce website, or a very extensive library, the common user may not think of that and the designer may not think of that on the surface level. But to be able to understand a subject and a topic and be able to break it down that way and know why you're breaking it down, by part or by type or by similarity, is very important.
Jorge: Why would a business leader, someone in a position of decision-making in an organization want to develop an ontology?
Ren: You know, it's really funny. So, a client many, many moons ago -- and this is, so ontology is a big buzzword, and within the information science community and the business world -- so I had a client who was a CIO, and the CIO brought me in to the office with some other consultants and said, "I want an ontology. And everybody else has an ontology, I want an ontology." And, I said, "That's great that you want an ontology. What do you want it on? What, do we need to focus on?" And the CIO said, "I don't know. I just want an ontology, you know, make me an ontology." And I said, “We have to kind of focus on something." And I kind of gave some examples of the business area that they worked in, and they just kept on saying, "I want an ontology." So, they paid an inordinate amount of money to make an ontology of pretty much nothing, because it was so amorphous, the guidance that we received, that we couldn't really focus on anything. So, I challenge business leaders to say, if you have a very complex subject where your employees, your clients, your stakeholders, can't wrap their heads around, then you might use ontological thinking. You may not need an entire formal ontology, but you may need to apply ontological thinking to that effort to be able to understand it.
Jorge: And the people who use those systems, how would they experience the results of the organization having gone through the work of producing an ontology?
Ren: Sure. So, the two most important things that I think would be the benefits of doing that effort would be one, being able to understand the things that you have within that project that you're doing, and two, being able to find things more quickly. Because you've described them and you described how they interact with other things, so you have an overall organization system.
Jorge: So, it's kind of like an inventory and also hooks for people who are typing things into a search box, or does it apply also to browsing such a system?
Ren: Both. So, if you're going to technically implement an ontology, yes, it would help your search engine to find things and would also inform your information architects and other people who are designing your website about where to place things and organize them.
Jorge: So, I'm hearing the phrase "ontological thinking,” and it evokes in my mind the phrase "design thinking," which is kind of an approach meant to extend the way that designers work to a broader audience. Is that the sense in which you mean ontology thinking?
Ren: A little bit. So, if we were carpenters, so to speak, and looking at this, design thinking, would be a chisel, and ontological thinking might be the hammer. Different tools, very similar in the approach that it gets a certain job done, and they complement each other, but they're a little bit different. And you would apply design thinking in a certain scenario and you would apply ontological thinking in a different scenario, but you might use them together in an overall project or endeavor.
Jorge: And as we were saying earlier, this notion that the paw is a part of the cat is something that I think comes naturally to many of us. What would ontological thinking bring to the table? Like, is there a particular framework that you teach folks to apply ontological thinking?
Ren: Oh, sure. Yes. And, I'll go to Jessica, who’s the one who coined the term. When she starts, she says, always think of the objects, always start with the things, the nouns. Try to find the nouns that you are working with within your project or endeavor and then go from there. And that's a very broad start, but yes, there is definitely a method or approach to ontological thinking.
Jorge: And in this case, the nouns would be paw and cat, right?
Ren: Sure. Paw, cat, living environment, food -- you name it, all the nouns that you can think of. When I do the workshop, I always apologize to people in the workshop that it's going to be almost like going back to your grammar classes in school, because once you find the nouns, then you find the verbs in between the nouns of how they interact with each other. So end up doing things like sentence diagramming, which a lot of people haven't done since, you know, grammar school.
Jorge: When you said, "sentence diagramming," it brings to mind a very particular way of rendering these relationships. And I'm wondering if part of this ontological thinking approach involves a particular style of diagramming, or are you literally talking about the sort of grammatical analysis that we did in school?
Ren: Yes, so for the workshop, just to kind of get people into the framework of being able to break down a sentence, we do very light sentence diagramming. But once you understand the subject, the object and the predicate within a sentence, never do it again. Just to re-spark that part of the memory of how to look at a sentence. But, on a day-to-day basis, after the workshop or when somebody does something like this... there are different ways of actually being able to represent the interaction between objects within an ontology.
Jorge: The image that's coming to my mind a is something like a conceptual map, right?
Ren: Exactly. Yes. No, the conceptual maps are what I call the architectural level representation of an ontology. So, for example, as you said, if you're designing an ontology for a search engine, you need to be very precise. So, I call those efforts more engineering efforts, where you have to put some good science and rigor behind the ontology, and there's tools to do that. There are well-defined languages to do that. But when at the very beginning, at the architectural level where you're applying more art than science, then yes, concept maps are great representations of ontology.
Jorge: Are there any tools or techniques that you have found useful to help folks? And we've already talked about diagramming sentences, but I'm thinking, that was in a workshop setting, where you're teaching people to do this. I'm wondering if in a business context, where you're actually trying to help a team or an organization create an ontology, are there tools that you use or techniques that you use to help them produce a first stab at it or something like start visualizing the relationships between the nouns?
Ren: Sure. So, card sorting is actually one of the basic, more natural, freeform ways of doing it and letting participants naturally try to organize these things. And then, that ontological piece. So, when they're organizing things up there, as you said, you know, people every day say, this is a part of this, this is a type of this. But I've found that when people do card sorting, they're not thinking, "Oh, this is a type of this, this is a part of this." They're just saying, "This term on this sticky is related to this term on this sticky." And then you come back later, when you're doing the actual analysis of the card sort and say, "Okay, these two things are related because there are types of, these two things are related because they're parts of." So that's a very good exercise for professionals when they are doing this for themselves. Mind mapping software is very good. Although, mind mapping software is very free form, so you have to build your own heuristic into the mind mapping software to make sure that you don't go off into just crazy, unorganized directions. I think those two methods, the card sorting and mind mapping software are probably the two places where I go and start.
Jorge: One thing that the conversation is making me think of... And I'm playing back to situations in which I've been in a consultant role trying to help such structures come alive or become visible to the people who need them. And one of the challenges I face in those situations is that in some ways, I'm documenting the current state when doing something like a card sort. It's like I'm mapping out the existing understanding. And in some cases, what you're tasked with is affecting some kind of transformation, or helping the organization move to a different place. Do you agree with that characterization, or is that something that isn't as big a deal?
Ren: No, I agree 100%, and that's one of the big things that when you build taxonomies or other information structures, often it's locked into a certain time. And when you say, "a type of," that means a type of right now, or is a, not was a, or will be a. And these are terms that information scientists normally use to link different objects together. It's normally just "is a." If you were doing a family relationship where you have a husband and wife, you know, if they're not a parent yet and you want to show, will be a parent, how do you reliably do that, in your design? And you know, that goes back to when I was an intern in healthcare, there was a huge problem with babies in hospitals, they call the baby logic -- or they called it the roach motel, in a certain sense -- because one person checks in, but two people check out when a mother goes in to have a baby. And the developers at the time were having a challenge of having that future state of... I have one person right now, the mother, but there's actually two people that will check out. So yeah, I definitely think it is a challenge.
Jorge: Are there any techniques to help effect that kind of transformation in the structure of the language they use?
Ren: Well, I think that you have to really take a look at your time component and really why you need to look at the past or the future, and then make a model of that and stick to that model so that you can represent the past, the current, and the future, and/or alternate states. You know, you may have different future states. So, it really, I guess, depends on the project that you're working on and it's just honing your skills as a information scientist, information architect, and being able to represent them reliably. I think in today's business world, we think too technically, you know. The T has taken over in IT, and we don't look at the information. And ontological thinking helps us to strengthen the information within businesses and their enterprises and really put a focus. And this shouldn't just be something that is done by information architects and information scientists and the data folks, but it should pervade into the business part all the way up to the CIO themselves and the way that they look at the information in their organization. So, I think that for the everyday person ontological thinking is, as you had said before, people think about this every day, and they organize things, but they're doing it subconsciously. So, if you actually consciously think of how does this relate to this? Is it a part? Is it a type? And what kind of part? So, you know, a paw is a part of an arm... That's kind of a physical part. But a remote control to a TV set is a part of a system, which is a different kind of part. So, if you think of things that way when you're organizing certain things, then you'll have a better understanding of what you're doing.
Jorge: Well, fantastic, Ren. It's been such a pleasure having you on the show. Where can folks follow up with you?
Ren: Sure. So, you can find me on SlideShare, and LinkedIn. Also, my company website, Info-Do, that's infodo.com. And I'm also launching a information set product where you have lot of the things that we talked about for information in mind mapping-type software and, and other type software. Specifically, we're starting with TheBrain platform and, that is at objectoria.com.
Jorge: Well that sounds really exciting. I'm looking forward to finding out more about that.
Jorge: Thank you so much for being on the show Ren.
Ren: And thank you for having me, this has been great.