Joe Gollner: A Multi-layered Definition of “Content” – Episode 205

Podcast: Play in new window | Download

Subscribe: Apple Podcasts | Spotify | Android | TuneIn | RSS

As recently as 40 years ago, we didn’t have much need for the word “content.”

But as soon as we started delivering the same information via multiple channels, we needed a way to identify the essential elements of content assemblies and to work with them independent of their various manifestations in information products. So the concept of “content” was born.

Joe Gollner has worked with content from its earliest days and has crafted a multi-layered definition to account for is many aspects.

We talked about:

• his work at Gnostyx Research and the research he’s doing for his PhD program at Lancaster University
• the origins of the “birth of content,” made necessary with the arrival of the ability to deliver the same information across a number of channels
• how to keep authors and audiences connected in decoupled content systems
• an animated dinner conversation he had many years ago with a developer around the intent of markup languages
• his distinction of the differences between content and information
• his seven-layered definition of content:
  1. terminological – content as that which is contained
  2. conceptual – content as potential information
  3. process – accounting for the variety of content roles in the creation of information experiences
  4. physical – content as a complex, composite artifact
  5. structural – content described at a simple, common base level, separate from the semantics about it
  6. organizational – content as an expression of organizational intent
  7. business/financial – content as an enterprise asset
• his desire that the content community convene a “meaningful, grounded, academically robust and business-meaningful” conversation around content so that we can bring more to our cross-disciplinary enterprise relationships

Joe’s bio

Joe is the Managing Director of Gnostyx Research Inc. where he specializes in providing objective and research-based guidance on the development, management, and strategic use of content technologies. In this field, he is a veteran implementer with over 30 years of experience, and he is well-known for mixing leading-edge ideas (and all too frequently concocting them) with highly pragmatic implementation tactics. He has Masters degrees from both the University of Oxford (Literature) and McGill University (Management), blogs about Content and Management, is working on a book about “Engineering Content”, and is pursuing doctoral research into the role of AI-enabled text analytics in the management of organizations.

Connect with Joe online

• LinkedIn
• Gnostyx Research Inc.
• Content and Management blog

Video

Here’s the video version of our conversation:

Podcast intro transcript

This is the Content Strategy Insights podcast, episode number 205. Forty years ago, we had little need for the word “content.” Books were books. Manuals were manuals. Life was simpler. But as soon as we started putting the same information between the covers of a book, on a CD-ROM disc, and then on the web, we had to sort out the content from its delivery channel. Joe Gollner has worked with content from its earliest days and has crafted a detailed, seven-layer definition of this foundational content-strategy concept.

Interview transcript

Larry:
Hi everyone. Welcome to episode number 205 of the Content Strategy Insights podcast. I am really happy today to welcome to the show, Joe Gollner. Joe has been around the content world for a long time and is very experienced. He’s currently the managing director at Gnostyx Research. We could literally talk for hours about content stuff, but today I really want to focus in on what are we even talking about when we talk about content. So welcome Joe. Tell the folks a little bit more about what you’re up to these days.

Joe:
Hi Larry. I’m really happy to be here. It’s a great way to kick off a day from British Columbia. I guess, in addition to my, we’ll call my consulting and integration work through Gnostyx Research, I’m also, terrifyingly, a doctoral researcher at Lancaster University in Northern England, and my research there actually does wind up touching, again, on content, no surprise.

Larry:
Yeah. So what’s your research on? We talked about this a little bit in Portland, but if you could remind for the folks.

Joe:
Yeah, I’ve been looking at, we’ll call it AI enabled text analytics as a resource, as a learning resource. And I’m very interested in using some of the technology to expose what’s in the content that people are creating and sharing. And I am particularly interested in how it impacts… What do you call it? … reflective group learning. How does a group or an organization learn? It’s partly about content, it’s about what do people articulate, what do they put in writing, buddy, but also, how is that internalized and made what people do? So that’s what I’m researching.

Larry:
Interesting. And it all starts with… I mean, maybe not starts with, but at some point, it manifests as content. But tell me, why do we even have to talk about content as a separate thing? We used to just have books and magazines and TV shows and stuff. Why do we even need this word?

Joe:
Yeah, it’s true. Really, one of my argument, I gave a talk several years ago called the Birth of Content, why did we start using the word content at all? And I’m old enough to remember the day in the ’80s when we didn’t really use the word content at all, but as soon as we had to publish information in multiple formats…

Joe:
So I remember in the military, we had great, huge printed manuals, massive multi volume monsters. And then, we started putting them on CD to save weight. So suddenly now, the content had to go in two different formats. So suddenly, we realized even the content is something different than what’s in the book or on the CD. It tells the two. It’s basically what we prepare to be able to print a really good manual or to create a really useful interactive, electronic technical manual, at least they call it, on CD-ROM. And then, of course, then the web came along, we went, oh, my god, now we have three channels, and so on and so on.

Joe:
And so, content became a term because suddenly, we said, “Well, when I open a book and start reading, when I go into a library or I watch a video, I am accessing content. I’m accessing some knowledge, some experience that’s in there.” And we talked about content as soon as we said, “But we have to manage that separate from any of the channels, any of the media in which it may be manifest.” And this is why we talk. And then we, of course, then had enterprise content management, web content management, and all sorts of manifestations and variations on the theme have emerged, and of course, with no agreement of what the heck it is that we’re actually trying to manage, which I think is a problem.

Larry:
Yeah. And still, as you say that, I’m reminded that I do a lot of content modeling for headless CMSs, but one of the big obstacles in that world is people still thinking of just like a web content management system and designing to a web page and seeing this one-to-one correspondence with the authoring interface and the end product. It sounds like you’ve been more successful at decoupling people from those experiences.

Joe:
It is something I approach with a good deal of respect. So yes, in the world of SGML, Standard Generalized Markup Language, the birth of angle brackets, although, technically, in SGML, you didn’t have to use angle brackets, but we’ll let that go. Basically, how do we mark stuff up in a way that’s going to be amenable to many channels, to omni-channels? Saying, “Well, this way I can transform it into that. I can do this. I can do this.”

Joe:
And in those early days, there was sort of a bigotry saying it’s all about separating the form, separating the structure, pulling things apart. And it never sat well with me. And I said, “It’s sort of true, but it’s also a problem.” If we abstract content too far from the experiences that it will be used to engender, one of the dangers is that when you’re writing it, you lose a sense of who’s reading it, why are they reading it, how are they going to experience this?

Joe:
So we call that a breakdown in rhetorical agency is what we talk about in the academic sphere is that authors suddenly feel disconnected from their audiences. And that’s the big problem with content is if we really are managing it as something abstracted from and separate from the many different channels through which it could be delivered, it’s much harder to get our hands on it and to really know, well, what is it we’re really working on?

Joe:
And this is why, and we’ll get to it in one of my layers of definition of the word content, but this is why, for many people, they never can make that separation. It’s always the webpage you’re working on. Or if you’re on the marketing side, it’s so focused on the experience and the measures of engagement that you’re looking for that this abstraction of trying to manage the content is quite separate from that is incomprehensible, they just don’t get it. Or developers, often, just don’t get it. They say, “It’s just something like I pull on this API, I get a bit of this, a few paragraphs, a snippet or some text strings, and I’m going to put it here, here, here, here,” that it is almost a bridge too far for them to figure out, oh, my God, I’m looking at this in the abstract. I’m removing it from the application experience I’m actually trying to create.

Joe:
So I do have a story. I was at a high-tech company in Silicon Valley and everybody there was literally half my age or younger. And basically, we went out for a dinner and one of the developers said, “Let’s put it out there, we hate XML. We’re developers, and we hate XML.” And I said, “That’s only fair.” So I went back to my soup, and he said, “What do you mean it’s only fair?” I said, “Well, XML hates you.” And I went back to my soup. And he said, “You can’t just say that and keep eating. What do you mean?” I said, “Markup languages like SGML and it’s sort of truncated offspring, XML, were designed to protect content from developers like you who, of course, you’re only interested in the experience that you’re creating de jour and you don’t care that there are other channels or other potential future uses of that same content.” So in fact, markup languages was born to protect content, human knowledge from the exigencies of developers hacking things. And oh, well, things got animated after that.

Larry:
I wish I could have heard that conversation. That’s brilliant. But it also gets it like, one of the things that we were talking earlier before we went on about our mutual friend, Cruce Saunders, and he talks about content as a valuable asset, and you’re like, “This is so precious, we have to lock this up in XML so that you can’t just do whatever you want with it because we need to use it for other purposes as well.” How did that conversation end? Hopefully, there were no fistfights.

Joe:
It was an interesting problem because of course the challenge that they had, and very often, with a customer, I’ll say, “We have to determine early on whether you have a content problem or whether you have an information problem.” And I said, “They’re related, but they’re distinct.”

Joe:
And if you’re very, very close to the experience or you’re very close to, we’ll call it, the governance framework in which transactions are managed, your problem may be predominantly an information problem.

Joe:
But if you are, say, sourcing material from a bunch of different formats and you have to get it into a common format from which to then engender two or more experience channels, then I said, “Well, then you really have a content problem.”

Joe:
And so, with that particular one, in turns out their job was a mix of the two. And we did find a pragmatic… So I am a pragmatist, so I’ve been called “Joe the Hack” at times where I will counsel a path saying, “Well, realistically, given your skills, given your objectives, given your timelines, here are some things that, basically… full content from other formats, get it into some kind of interim working representation.” So we’re kind of like short-circuiting things. All they wanted was the experience. And so, it worked out.

Larry:
Can you elaborate on that distinction between content and information? Because I think that’s one of the first things I like to tease out when I’m talking about content to distinguish it from other stuff that we’re working with. How do you distinguish those?

Joe:
This is a great way, a great question to start into my seven layers of definition. And people out there who know me will be rolling their eyes going, “Oh, here goes Joe now.”

Larry:
That’s why I invited you, just by the way.

Joe:
The first level, which we call terminological, is basically back to the Latin root of the word content, which basically is that which is contained. And I said, no, Canadians might be thinking about jars of maple syrup, the contents, the maple syrup, but I said, “But we’re not really talking about…” But if we’re talking about information, it’s like the content inside the information. So we talk about information content. And that can be very broad, it can be the video content, it can be whatever, the visual content. Very often, it’s the textual. And even where we have multimedia representations, frequently, there is a representation like a script saying, “This is what we were trying to do with the video,” or whatever. So that, we call it the terminological level, it’s very, basically, content is that which is contained, and specifically we’re looking at information content.

Joe:
And that leads us to the second level, which I call my conceptual level, which is if we follow that to its logical conclusion, then we are compelled to define content as, quote-unquote, “potential information,” it’s the content within an information transaction. And specifically, it’s the content that can be, actually, delivered in multiple, different information transactions. So it could be…

Joe:
And information transaction is a very, very selective term. I define information as the meaningful organization of data that is delivered in a specific context and with a specific purpose. So specifically in the word information, inform is about guiding or influencing the behavior of audiences. So that gets us to, we call it the transactional, the experiential side of information. And I said this is really useful because then content is the material we put into those messages, those transactions.

Joe:
So it’s very difficult. So by saying it’s potential information, we’re not really separating the two, we’re saying that content is intrinsic, it’s internal to what we call information, but information is an action. So it’s like Speech Act theory. When you talk about information management within organizations, it’s all about the responsibilities associated with information activities, so governance, liability. It’s been always enjoyable to deal with the legal part of organizations because they have a very, very concrete sense of the liabilities associated with information. And that acquaints us with, it’s an action. If you say something, you will be held to account. It’s like whatever you say will be held against you in a court of law.

Larry:
That legal term of, “is it actionable?” That makes more sense to me now in the context of what you just said. What’s the third? So after that, the terminological, the conceptual, then?

Joe:
Yeah. And certainly, with that conceptual one, as soon as we make it intrinsic or bundle it with our concept of information, then we can look at data, of knowledge, and it can all fit together.

Joe:
And perhaps suited to a different talk is what I call the knowledge dynamic. How do these things actually interact in practice? And what role does content play inside there? And basically, as potential information, it plays, we’ll call it a linchpin role in the knowledge dynamic.

Joe:
But the third one, if we move to the third one, actually, we’ll call it the process perspective, where we think about the roles. And it was seeded by one of your points, Larry, about we have people who work on content, we recognize they’re the people, the strategists, the planners, the designers, authors, editors, translators, animators, videographers, whatever, people working on the content. And I guess we include in that the people managing, curating, publishing, those are all roles where people get their hands dirty, a bit like the Cassandra Nova in Deadpool 3, getting your hands dirty in the content.

Joe:
And then, we have others who, almost like, they’re really much closer to the information representations or transactions that can then be engendered from the content. So really, for example, UX designers, developers, people who are very, very attuned, maybe marketeers who are really close to the experience. They really are fussed about, what’s the impact on the audience? Am I getting them to do what I really want them to do? So they typically have a very different perspective on content.

Joe:
They’re much closer to the trend, and that’s where I often have to raise the question, are you really interested in the content or are you really interested in the information impacts or experiences that are being engendered? And this is an area of frequent tension between perspectives. But once I put it out there and paint the picture that way, people say, “Oh yeah, I do see it from a very particular perspective, and I can see that there are other people who have to look at content in a different, broader, more longterm perspective.” And we can sort of agree to disagree in sometimes. Yeah.

Larry:
Yeah. I love that articulation of the process that it seems like a pivotal point of this. Anyhow. What’s number four? Now, I’m dying to hear the rest of your-

Joe:
Number four is physical and is the physical characteristic of this thing called content. And then, infuriatingly, I define it as a complex, composite artifact. And in Joe fashion, this is a very, very complex definition, but particularly, it tells us content brings a lot of things together into a very, very impactful manifestation.

Joe:
And so, partly, this aligns with saying, okay, well, what do I bring together? I say I bring some data resources together, I bring data together, I bring links or associations with knowledge precedent. Like what are the key points of reference that I’m connecting to? I bring in sort of a foreknowledge of the multiple audiences I may be engaging, in the impacts I want to have, but they all have to come together in this very complex artifact.

Joe:
And by composite, it’s both internally representative of the fact that it has all these moving pieces. It’s actually a hyper structure or hyper sophisticated artifact, but it also, as a composite, it builds with other ones. So sometimes we talk about microservices and microcontent for microservices, small, little nuggets. And I said, well, true, but very often, the meaning of, we’ll call it a component of content, is dependent on its place with other ones, it’s composition with other ones.

Joe:
And that got back to my saying, if you try to overly separate content from the experiences, then it can feel less alive, it can feel less real, it can be less impactful. So I have projects we’re saying, while we’re getting things approved, we have to actually aggregate it into certain types of artifacts like the approval or regulatory process operates at this level, but then, in my later days I have the components which I can then use in experiences, but it can always be connected back to the representation that was approved as its authority comes from this composite level.

Joe:
So you can use it independently, but you can always track back to where did it get its authority. So in things like medical or aerospace and defense, we have a very refined sense of authority. Why are you allowed to look at this? Why are you allowed to do this? Why must you do this? And that comes from, usually, higher levels of content. So the physicality, so that’s that fourth level is the physical representation.

Larry:
Nice. I love that. It reminds me of, in my content modeling work, I have this thing I’ve always called an assembly. I had the chance to compare notes with Noz Urbina recently, and he calls it an info product. And we’re both talking about that same thing that I think that you just described right there, the physical manifestation of this thing that’s available for multiple uses, but it’s sort of the authoritative, definitive information chunk that you intend to deliver.

Joe:
So a phrase, information product, is one that we’ve used for 30 years and more. And they differentiate, I have these content assets, they get manifested as information resources. Each of these terms is chosen with profound deliberateness. And they come together in information products that engender information experiences. And there are transactions for which an organization is held accountable. So it is a really important thing.

Joe:
And to assembly, for example, if we go back to a standard that I wrote, sadly, back in the ’90s for the military, what, in DITA, we would call a map, we called an assembly. And you had multiple assemblies for pulling together all the content assets together in different views. So some were by information product, first line maintenance manual, second line maintenance, manual troubleshooting, manual, whatever.

Joe:
But also, you had multiple breakdown structures, you had the functional breakdown structure, the physical breakdown structure, spatial breakdown structure, and all of these became assemblies, ways to access and navigate the content. So it’s kind of like an early company that I created was called Zia Information Architects, and we were interested in the architecture of information, really it’s architecture. And of course, then information architecture became popular and the web not about that. So it was kind of like the superficial baby version of information architecture. So I changed the name of the company saying, “I’m not even part of that, it’s irrelevant.”

Larry:
Yeah. What’s number five? I’m getting more and more curious the deeper you go.

Joe:
Five is really a computing computer science perspective. And this is where I’m actually, sadly, applying a lot of my energy these days. It’s like, okay, well, from a computer science perspective, what is it that we’re really interested in with content? And I would argue it’s the structure. I said there is a root or a base semantic that is inseparable from the structure of human communication. And sometimes it’s called the publishing semantic, but it’s the structure, paragraphs, lists, hyperlinks, tables, visual like figure reference or whatever it is.

Joe:
We have a structure. And it is extremely common. If you go across S1000D, CAL, DITA, DocBook, that’s encoding initiative. If you look at all the standards at the end of the day, you can boil them all down like maple syrup and say there is a structure that humans give to their information, their content. And this is always common.

Joe:
What we still don’t have and we’ve been working towards for decades is a sound computational approach to the structure of content. Part of our challenge has been we have often muddied the waters between that structure and the semantics we would like to then apply. And I am firmly in the camp now saying, following Ted Nelson, we want to separate the semantics from the structure because we can have multiple semantics applied to any content structure. And the semantics are really about applicability, to what does this apply? And you can take that down to the tiniest conceptual level or to a very high level business entity.

Joe:
So computationally, then, we say, well, content needs to be validatable, highly processable, it needs to flow, runs application, it needs to move between multiple applications. It has to be easily represented in multiple, different formats. So I sometimes call it REST-ful content. It can genuinely be represented in different ways.
For example, if we look at our parsers today, and believe you have, and they call it back parsing like with DITA, it’s pretty clumsy. It’s like, in some ways, we did this better in the ’80s. That’s bogus and sad as Wayne would say in Wayne’s World, it’s just wrong. So computationally though, it has a very specific portability and multi-representational transformations. That’s really what content needs to be.

Larry:
Interesting. As you say in that, you’re reminding me that there’s such a diverse ecosystem of tooling for the computational management of content. Like every CMS vendor in the world has an opinion about how it should be done. There are standards like DITA that drive some of it. I guess, what are a couple of generic top level ideas that people can take away from that insight, especially when you’re content modeling or doing information architecture, to think about how to keep it in that kind of computable model? Does that make sense?

Joe:
Yeah. Yeah. So when I do look at content modeling now, I tend to be quite harsh and say, at the root level, there’s a common structure and that’s almost like a given, you can almost extract it from all the preceding standards.

Joe:
I would go further and say the time has come for us to do a consolidation between SGML, XML and what we tried to recover from SGML and DITA, we had accidentally thrown away in XML, there’s a base level and then the semantics get very precise. So the content structure actually becomes quite generic and tuned to flow. It needs to move between many applications. And many people working in the content space are working on a particular CMS or a particular one or two channels, but they haven’t worked in, we’ll call it, the real enterprise zone, the vendor zone where your content’s going to move between 10 applications, highly specialized applications.

Joe:
So anybody who’s experienced, moved content through multiple applications, and what I call a donut of equivalence, will be chastened quite quickly. So the content layer becomes very, very simple, almost universal. And then, the semantics become highly precise, and oftentimes, managed separately. So represented in JSON or in a graph knowledge base or something like that saying it’s over here, and that we bring these two things together in multiple ways, including, if necessary, expressing the semantics using our markup, but bringing it together in a way that it allows multiple semantics to be applied to the same structure.

Larry:
Yeah, interesting. I’m revising my content modeling thoughts as we speak, but okay, number six. Got it.

Joe:
Number six, it’s called the organizational or organizational learning perspective. And yeah, that’s where we talk about, also a big emphasis for me is that content is really an organizational expression of knowledge and in intent. In other words…

Joe:
And this tells us something really important because content is not so much what you or I or any one individual has to say, it can be, but when it’s interesting it’s when the organization has come to agreement and said, “As an organization, this is what we say on this topic.” And that’s what makes the work that content professionals do so hard. They got to go get legal approval and they got 10 subject matter experts who really don’t like them and so on and so forth. And they hammer out an agreed articulation, which then…

Joe:
And this leads us to the seventh one, which is really called the financial or business perspective, which is when you have that organizational expression of knowledge and intent that it becomes very, very valuable. It’s only by going through those six preceding levels that you can really talk about content as an asset, as an enterprise asset.

Joe:
So I said content can be an asset, it can. And accountants tell us very clearly what asset means, it’s something with the potential to deliver future returns, which gets us back to our original concepts of content is potential information. It’s like we can use it in many ways. So if we’ve gone through the first six layers, then we can genuinely talk about content as an…

Joe:
And so, I’ve debated, I’ve had great fun with Cruce, for example, about saying, okay, well, content may be an asset, but it actually has to get there. It has to be managed in a specific way to earn the title of an enterprise asset. But really, if it’s done well, that’s exactly what it is.

Larry:
Yeah. You’re making me wonder now if these layers of concern could be the foundation for something like a maturity model. I think that might be overused at this point, but like an organizational assessment, like how far along are you in really honoring your content?

Joe:
Yes. Yes. And even this year, I had a customer going, “You have a maturity model?” And going back decades, we’ve had so many maturity models. And sometimes, they do strive to be a real maturity model, but normally, they’re like, how much do you love DITA, for example? Well, that’s not really a maturity model, but certainly, yes, you could structure this, and I’ve taken a run at it a few times, but certainly, I think all these seven perspectives do have a role to play.

Larry:
Yeah. And I love that this is going to be food for thought for a lot of people, but Joe, I can’t believe, we’re coming up close to time, but is there anything last you want to say, anything, any way you want to wrap up the list or anything you want to make sure you share before we close?

Joe:
Like this discussion, I would love to see, we’ll call it a community or find a way for people to come together and really start to engage. And there are multiple perspectives. I tried through these seven to sort of bring together the different camps, but I think it is a very useful concept. And language terms are tools and you rightly evaluate them on whether or not they’re useful. Do they do any good?

Joe:
And oftentimes, I’ll hear definitions of content saying, okay, well, that’s fine, it doesn’t help. And so, I do encourage a discussion around how can we come to a meaningful, grounded, academically robust and business-meaningful usage so that we can engage stakeholders. Because one of our challenges always in the content business is how do you engage the bean counters, the strategists, the IT groups in a way where they say, “Okay, I see where you’re coming from. I see how you fit into the picture. You came to play, you got some cards, we can deal, we can play.” And very oftentimes, the people in the content world show up with nothing to deal, nothing to bid with, and we can do better.

Larry:
Yeah. And I feel like just considering your list is going to help a lot of people in those, because like you said earlier, we’re just always fighting for that seat of the table, looking for the respect and the value in the org. I personally feel much better equipped now, so thanks. Hey, one very last thing, Joe, if folks want to connect with you or follow you online, what’s the best place to find you?

Joe:
Oh, well, probably LinkedIn. I’m easy to find on LinkedIn. And I actually have a blog, www.gollner, my last name, .ca, so the Gollner in Canada.

Larry:
Got it. Okay. And I’ll link to those in the show notes, of course, as well. Well, thank you so much, Joe. This was awesome. I really appreciate your perspective on this.

Joe:
Well, my pleasure, Larry. It’s been a hoot and you can see this has provoked me. Yeah.

Larry:
Yeah, no, I can see why you’re in a PhD program.

Joe:
Mm-hmm. Yeah.