Small Talk with João Rosa and Trond Hjorteland (Transcript)
This blog post is the transcription of the chat we had with João Rosa and Trond Hjorteland about their Intentional Architecture Workshop on 19 May 2023.
The conversation has been slightly edited to better fit the written format. Enjoy!
Avanscoperta: This workshop is called Intentional Architecture, a very promising title, and I'd like to start from the basics.
What is an intentional architecture?
João: Last year, I was in Barcelona at a conference, and I actually had dinner with Mark Richards, the author of some books from O'Reilly. We were talking and I said that we don't have enough intentionality with our architecture, and he open his eyes, saying, “Yes! I need to grab this concept!” because that’s what it is - we don't have enough intentional architecture.
And this is when the concept starts to take shape. Everyone can say that, yes, we have intentional architecture. But I could challenge everyone. Do we really take into consideration the environment around us? Do we consider the actual state of our company and what is able to do?
Or are we just looking at the technical components and say, “Well, because we are using Kafka on AWS, we are being intentional”?
So when we say intentional, we don't mean the technical components, but rather how do we do architecture and design to meet the human needs. Are we really being intentional in that direction? And that is how the name pops up - intentional architecture.
In our workshop, we're going to take people on a journey on how to scan the environment, how to look at the current capabilities in the company and how to generate these options to keep, evolve and be a sustainable company where people are happy to work.
Avanscoperta: So today, we are talking about team structure and team organization, rather than only the technical aspect.
And one of the things that we see in the workshop description is to "put the people at the centre of the design” somehow.
Were there any other moments where you felt we need to make this kind of decision more intentional? Why intentional in the first place, as opposed to non-intentional? So you mentioned before that the focus was on the tech only.
João: Let’s say intentional VS accidental, right?
When you just look at a portion or a fraction of what people in charge of designing and architecting a system should do, lots of decisions are accidental, and some play out, and others don't play out.
What we try to bring here is the frame of thought to analyze more what is around us, analyze more around the environment and start generating options about the consequences.
For example, if we use x technology, this has a consequence for the teams. Or if we don't have y skills on the teams, how can we acquire these?
But we also want to do this in a collaborative way, rather than imposed on top-down, because everyone has their experiences, and how we perceive the world. And really, we can do this together, we can do this in a different way that is not a silo, is not an ivory tower, and this is what we're going to explore in the workshop - how to have intentionality when we have a group of people because it's very easy.
For example, I'm starting my own company and I'm doing software alone. Right now, I’m making all decisions, and it's different from having a company of 20 or 100 people.
And there are lots of practices that actually come from social sciences, which is from where we get our inspiration for the workshop too, and we can embed that in the software industry, as well as good practices that we already know and that are already being used in our industry.
By mixing this together, we hope to create this intentionality to evolve architecture, to evolve organizations where humans are at the centre, rather than having the machine and the software in the centre… we want to put humans in the centre.
Avanscoperta: Let’s take a look at what is in the description, sometimes some words might sound a bit obscure and we talk a lot with jargon.
When we talk about open socio-technical system thinking, what do we refer to? I’m guessing it’s about team architecture and team structure, and you just mentioned something important - social sciences are a big source of inspiration.
So what is the context of open technical system thinking? What do we mean by that?
Trond: Yes, as João mentioned, this comes from social sciences predominantly. And this is where I spent most of my time, at least during COVID, because I started reading materials and papers, with plenty of time on my hands, and that’s how I got into it. João and I have natural science, IT, or STEM (Science, Technology, Engineering and Mathematics) background, or whatever we want to call it, so this was a new area to explore.
And just to bring it back to the title of the workshop, which is Intentional Architecture, I actually pitched it to one of these practitioners in the social sciences on this, and they said, “Oh, that’s cheeky!” It’s a different name to purposeful systems because that's the term they use in social sciences, right? So they see organizations as a system and then you have to align the purposefulness of not only the people but also the organization as a whole.
In the IT and technical space, you employ a lot of technologies, not only IT-technology but also the techniques that we use with the people. So we're referring to techniques and technology in a broader term.
And what we see is that you can't optimize for either of these. You have to optimize them jointly. You have to look at both of them at the same time.
Human relations, for example, focus a lot on the social bit, with concepts such as giving a pay rise and providing social security, or doing a course on how to communicate better and all that stuff. This means you’re focusing a lot on the social bit.
While others might focus too much on the technical bit - this is what they discovered in the social sciences, going back to the 50s, which is when they introduced new technology...
if you ignore the social aspects of it, the technology is not going to perform as well as you would expect it to do.
So back to what João said earlier, you have to bring people back into the system. You're not going to throw away the technology, of course, but you have to view them jointly. And that is what a socio-technical system is. Which means we’re talking about teams and people in the context of an organization.
Let’s talk about the “open” bit, too, because socio-technical system design is like a term on its own - STS theory.
We added “open” because there is a branch of socio-technical system thinking that goes in that focus on systems that are open, which means that they are exposed to the environment. And the environment can be your industry and your competitors, but it also is the wider social feel, such as how people feel, and how they experience the world. I mean, people come to work as a whole person, they don't come to work as a machine part fitting into that organization doing exactly just that.
So we have to realize that every system is generally open, and the open social system thinking is just taking care of that as well, not just closing it and creating what a colleague of mine here in Norway call “a terrarium”, which you really can't create, that's unnatural. You're going to limit your system by closing it. So just keep it open.
Avanscoperta: So we’re putting together some things that already existed both in the IT and technical world and in the social sciences world, aiming to create something new.
Who is this workshop for? Who is going to get the most out of it? I guess it might apply to someone who has some kind of power or has to make choices in their company, is that correct?
João: This workshop is for people who design systems and design socio-technical systems.
So depending on your company and the titles that you have, it can be a software architect, but also a senior engineer or a principal engineer, it can be engineer managers, directors and even people from HR because HR policies can constrain how we’re gonna structure our team.
So the workshop is for this type of people, but especially
this is for anyone who feels that something is not quite right.
I've been talking with some people and part of my consultant work is also being interim CTO - I'm building new teams in a company, so interviewing and building first teams. I'm looking for senior engineers who are uncomfortable. And I interview very interesting people who say, “I excel in what I do, as you can see in my portfolio. But I feel that something is wrong”. And I always go, “Tell me more”. “The way that we interact is not quite right, the solution is not there… ” etc.
The workshop is really targeted at these people who are good at what they do, but also know that there’s a lot more than that. But they don't know what this “more” is.
So this is what we want to explore as a group. “More” can be the “open” part that comes from social sciences, so we’ll want to learn how we get these signals. How do we learn from the environment? How do we plan to be intentional and change the architecture of our software?
So we are looking for these people who are uncomfortable with the current status quo and think that there must be a different way, that there’s is something that they’re missing.
I went through this alone, and as Trond mentioned, I also have a formal IT education. Everything is very technical - how to normalize the database, the patterns to code, etc. And then, inside of me, I started to feel that something was wrong, that something was missing. So I start this journey as well, asking myself, “What is missing?” And so, with many influences from Trond, we realized what was indeed missing.
And that’s why we created this workshop. This is for those who have these responsibilities and feel uncomfortable.
Avanscoperta: We can say that we are aiming to provide a new, more comprehensive way which takes into account the whole human aspect for people who, as you just said, want to change but maybe don't really know where to start.
There are three main topics, namely Domain-Driven Design, Team Topologies and Wardley Mapping, that come into play into this because all of them put humans at the very centre of their actions.
How do relate to each other, and how can they help in this? How are we going to use them during the workshop?
Let’s also highlight that this is not a workshop to learn Domain-Driven Design, Team Topologies and Wardley Mapping - just to make it 100% clear.
Trond: The thing is that Domain-Driven Design, Team Topologies and Wardley Mapping can work perfectly, and if you hit the right wave, they are spot on, right? Everything is going smoothly.
Domain-Driven Design means everybody is engaged, and you're creating fantastic solutions. You do Team Topologies to have these fantastic teams and with Wardley Mapping, you do a super strategy… but a lot of times, this doesn’t work, as simple as that.
So why sometimes this mix works, and other times it doesn’t?
It can depend on a million reasons, but what we try to attempt with our workshop is bringing social sciences in so that we can explain why DDD works here but didn't work so well there, and why the streamlined team fell apart in this company but worked very well in another, or again why Wardley Mapping helps some people creating amazing products but is not helping some others?
So the point is bringing the social sciences in and understanding these techniques better, and also tuning them a little bit to make them work better in different contexts.
João: And there is a fourth one (other than Domain-Driven Design, Team Topologies and Wardley Mapping), that's the Easter egg, which is what puts everything together, it comes from social sciences… but we won’t reveal it now as it might scare people away.
The first stage is learning about the environment and creating some action plans, and how to do this as a group.
So when people ask themselves, “How do we do strategy?”, people think that strategy is done by executives in a hermetic room, or that people put proposals in a mailbox, these get picked, and this is the strategy. That’s why strategy fails.
And lots of us, in the IT field, think this about strategy. They/we think that strategy was this dark magic that you need to evoke from ancient ages… to do strategy.
Actually, it's very easy when we bring in the social sciences - and here is where Wardley Maps fits well, because Simon Wardley created these visualizations that help us with questions such as, “How can we analyze the environment?” because there are different ways to analyze the environment.
We will use different ones, and Wardley Maps is very good at doing that - “Where we are?”, and, “Where can we go?” So the body of knowledge is perfect. We are just putting the Why. And this is the first part.
The second part involves using Domain-Driven Design and Team Topologies, where we’re in the stage where we have an action plan, and now, “What are the things that we need to do?” Maybe we need to change the scope of teams now that the software needs to evolve.
And we find that in this body of knowledge, they have something in common - they are holistic. So they look to the whole, they don't try to partition everything.
But especially with software, we need boundaries. Especially when companies grow and software systems become more complex, if we don't have proper boundaries and isolate language, we know where we end. That's why now Domain-Driven Design and its techniques are picking up, and also Team Topologies gives us a pattern language, which is exactly what we were missing to talk about team types and, most importantly, about the interactions between teams.
“When do we want to use an interaction that is collaboration?”, or, “When do we want to do X as a service?” are key questions to ask, because they have implications, we just cannot throw things.
As Trond was saying, lots of people use Team Topologies because everyone wants to have a platform, plus it’s a new thing, so folks want to try it… but “How do we design platforms?”, “Where are the boundaries of these platforms?” We will collaborate to discover boundaries and what happens when we move to X as a service.
This is all the part of being intentional that is missing.
Therefore, we use this Easter Egg technique I was referring to, the one coming from social sciences, to give this frame and explain why that is what is missing, and then we bring techniques and the principles of the IT-based techniques from our industry (DDD, Team Topologies and Wardley Mapping), these bodies of knowledge that people are used to, and in doing so we will glue everything together, to have a coherent journey where people start to see how can we be more intentional.
Trond: Just a little note. It seems like we’re bringing in techniques and insight from social sciences into IT and building on that. But there is also a flow the other way around because social scientists are really interested in what's going on in IT. Some folks in the social sciences field they're really curious.
So I believe that using these techniques, actually the three we mentioned here (DDD, Team Topologies and Wardley Mapping), we can actually maybe even affect how they see things from the social science perspective.
How to do strategy, the Domain-Driven Design part, and how to actually create a congruent organization to the technical bits - it all fits together in a nice and beautiful way.
Avanscoperta: And as João mentioned before about strategy not happening in a dark, secret room, no change happens in a bubble, right? All the fields of knowledge somehow influence each other, so it’s understandable that IT and social sciences can exchange ideas and contributions along the way.
And we’ve noticed there’s a lot of interest in the IT’s culture of work and teams. From the outside, it might not seem so, but the complexity of the topic is huge and our industry has done a lot on this topic. Thanks for pointing that out.
Now, we briefly mentioned this workshop is taking place in person. How do you plan to structure these two days that we're going to spend together when it comes to theory VS practice?
Trond: I can say that the whole workshop is going to be predominantly practical because we don't only bring in the three approaches that we already mentioned, but also three core techniques from social sciences as well.
So we are melting this together and it's going to be a hands-on workshop altogether. But what we're going to do is that we're going to introduce theory as we go along because all these practices from social sciences are infused by theory. They are based on a solid theoretical background.
Why are we doing this in this order? Why are we doing this before the other? Why are we what's the structure of this thing? All of it is informed by theory, experience and experiments. It is sort of funded in a solid theoretical framework. So I think that's going to be interesting.
Also, as I mentioned earlier, we also can use that theory to explain things that we see in the other techniques such as Domain-Driven Design, Team Topologies and Wardley Mapping.
Also, I want to point out that I'm happy that we are able to do it in person, because some of the social sciences techniques we’re gonna be using are funded on people being in the same room because it's a social interaction thing. And you don't get that with a 2-D remote scenario.
I would like to experiment a little and see how far we can push it as we did during COVID. We pushed EventStorming quite far (how to forget? ed.), so it could potentially be done… but I'm really happy to be able to be in the room together this time around.
Avanscoperta: Another question we always have at Small Talk is, “What kind of people will benefit from this?” We already talked about this briefly, so I could expand by asking, “What companies would be the best fit for it? Would it be medium, small, or large companies?” Any thoughts to share about this?
João: Any size would be the easy answer. But for this to really work - we’re talking to companies that know that the traditional methods that have been used since the 70s and 80s are not working.
And actually, as a consultant, I work with a telco (telecommunications company) that has more than 3000 people, and I was working with C-levels on helping them to find new boundaries and ways of working to move forward. And their structure was the one of a classic consultancy company.
So for this to work we’re aiming at companies that know a lot about some techniques, and they use an approach which says that if you did X then you would get Z - some kind of scientific management principles. These companies are our target.
But this workshop can really work at different scales. For example, we might have a CTO of a small-medium company joining in, and it could benefit them too.
The point is, as I said before, that these people need to be uncomfortable, and also the organization is uncomfortable, so they’re wondering, “The traditional way of working is not working, what is out there for us?”.
And rather than the size, we want to focus on these companies that are feeling this pain, because these companies are the ones that are going to change the world, and they’re the ones that want to change the way they operate and, therefore, the way they will attract new talent.
Let's be honest, although we read about many layoffs, there are still more jobs than people. So we as individuals have the power to choose organizations that align with our ideals. If I like honesty, if I like transparency, if I like to be involved, I want an organization that matches with that, so today organizations are also slowly turning to these values to empower people.
Can we support people to have better well-being and less mental illness? All of these types of things are connected, and this is what we want to explore.
So we are looking more for this type of company that is already primed for this change rather than the size itself.
Trond: Companies that probably would hire management consultants to do their bidding, like McKinsey or whatever, probably wouldn't be our target group here.
Our target would be a company that believes in its own employees and want them to be engaged in work and be happy workers. And they are ready to participate in the design, whatever that is, even if it's the strategy, or down to the team level.
It has to be a company that has trust in its employees.
Avanscoperta: You already mentioned there are people who are uncomfortable with their situation in the company. The point is also that you can’t convince people - they need to realize there's a problem by themselves. You cannot come from the outside to promote change.
Still, how can we help this happen? You need to feel some pain, as we said.
How could we say to someone, “This workshop is for you”? What problem are we really aiming to solve with the Intentional Architecture Workshop?
João: You have your software architecture, you have your teams, but you constantly see your time to market increasing. You try different stuff, but because the environment is constraining you, nothing is working. So now, after all this trying, the managers are in a better frame of mind to want to try something out of the box and do this experience with us.
So this is the type of people we’re talking to, those who are at this inflection point. Some companies are already figuring this out for them, and they have generative environments. If we look at the income, they were always being diverse since the beginning, and because they were diverse, they figured out these problems. Some companies are already figuring out that.
We are looking for these companies that are trying to solve these problems, such as: Can we have a better time to market? Can we attract people by having an inclusive environment where people can do designs for things that they like? We are IT people, so we like these problems and how to solve them, but can we also empower people rather than have top-down management?
So we are trying to show people that there is a different way of doing things, and if you want to convince your manager, join the workshop. We are also more than happy to go to your company, but join the workshop to have the experience, bring another buddy to do it and then experience and say, “Oh, maybe this can help”, or, “No, thanks for sitting the idea”, and maybe in one year this will work - we planted the seed.
These are the types of problems people are having, these mismatches in designs, all of these types of things… because the bodies of knowledge that we were talking about, such as Domain-Driven Design, well, this is nothing new. Eric Evans’ book came out 20 years ago, and he was already doing it. And the new element is bringing in the social sciences.
Let’s ask ourselves, what is really different between software and an earphone case? Both software and this case are artificial, right? They are products created by humans. But software has a different quality because it is very easy to change in production, and it is immaterial, while a case made of plastic - I need to destroy it and go through the production process to create it again.
That's why we see industries powered by software going faster and faster, which is why we see the world going around so much faster.
The big difference is between manufacturing companies that already use technology versus pure digital companies or companies that leverage digital. I know this from my experience because I used to work in manufacturing during summer time to make money, and moving a 100-tons machine is hard, right? You don't move it easily. But changing software is different, it’s easier, and it can also have big consequences and big outages.
But this is the difference, and this is what we are trying to solve, answering questions such as, “Can we have a better time to market? Can we eat our outcomes? But at the same time can we have a culture where people feel safe, feel included, and design is made by people that are actually building it, and not by someone far away?”. This means that part of the strategy also needs to come from these groups of people.
This is what we are trying to solve. We know that it is a lot, we are not saying we're going to solve this problem for people that have problems with streaming events and more classic technical challenges.
We are going a step up and this is why we are aiming at people that feel uncomfortable because they tried many types of architectures, they moved technologies… and it didn’t work. Yes, it didn’t work because not all is about technology, there’s more to consider.
We want to get back to this human-centric idea where people that are building the software are the same who make design decisions and are also the same making strategic decisions.
It is not scary, this is not too big, because we as humans are responsible, and when we are given responsibilities, we will act.
Avanscoperta: Every workshop starts with someone having some kind of pain, as bad as it sounds, and the workshop helps resolve those problems.
Now I’d like to ask if you've had any case studies or some instances where you've seen this in action or working somehow. Has anyone already experienced this workshop in some shape or form?
Trond: From my knowledge, this is fairly unique, this is the first time I've actually seen these elements put together in sort of a melting pot.
As mentioned, social sciences have been doing this in various industries since the 50s. There are examples of Telco and some companies in Australia too, and many more from around the world. Most of them are in manufacturing, where they’ve been using lean for a while. And even this type of company realizes they need to have a more human-centred and team-focused way of organizing the work.
And in IT, we have Agile. So we have been trying to do this for quite a while, but we are now bringing this in to empower Agile somehow, making it more profound and also introducing the techniques mentioned earlier.
João: As far as we know, none has been adding the social science element to the already-mentioned three techniques and approaches (Domain-Driven Design, Team Topologies and Wardley Mapping). There are people teaching these three things, of course. But no one has included people that study humans and study groups in the conversation, for what we know. Please, if there are, do tell us - the world is very big, and we might not be aware of it.
In terms of case studies, I'm about to publish a case study with the folks of Team Topologies that is around this - how we helped a company move their operating model. I was there supporting them in doing so. We did this with their people and used these techniques to reason about where the boundaries will have to go because they have a typical demand and supply type of organization, with the business and the IT. It is the first organization of this size that I know that moved to a capability model. Let’s say there is a capability, such as self-fulfilment of broadband. And this capability has a process that has people that are oriented to process, and there's technology, right? Order management systems, tracking systems and stuff like that.
It's the first company putting this together under the same department fulfilment capability. It will be published in a few weeks, and it’s a nice case study where people can understand how to bring this together and put humans in the centre.
The company was on a straightjacket because the time to market was impossible, and the need for coordination across the company was gigantic.
To add to what Trond was saying, IT and social scientists have already worked with two well-known companies in the past - Microsoft and Cisco, so this has already happened somehow. Things didn't turn out that well for Cisco, but Microsoft was able to reinvent itself a couple of times even when everyone thought that they were gone.
We are just trying to democratize the whole topic and expose more people to the thinking to try and see if it goes out.
Avanscoperta: Can you suggest anything in terms of readings, videos, or books for folks who want to start studying the subject before the workshop?
Trond: There are some blog posts I wrote as I started learning about this topic, it’s been going on since 2019. These posts are part of my learning journey, so the first ones might be naive. Anyway, I believe it can be a good place to start, at least for the social sciences bit.
Anything about DDD, Team Topologies and Wardley Mapping has already been covered to a great extent.
João: Trond’s talks at DDD Europe and KanDDDinsky are also a good place to start from.
Trond: There are some books in the social sciences field, but to be honest, they are tricky. It's not an easy read at best. Also, there are loads of papers, some are easier and more accessible than others.
On that blog that I mentioned, I've listed a few of those papers that I recommend people to have a look at, but it's not an easy entry, any of them.
So, yeah, somebody should probably do something about that at some point. For the time being, let’s just refer to my blog posts and what you can find on LinkedIn (see below, ed.).
On my blog, there is one blog post that is actually listing a few of the reading materials.
Avanscoperta: Thanks for that. So I left the best for last, as they say. João and Trond are actually publishing a book on all the topics we’ve covered today. Can you tell us a bit more?
Trond: We haven't signed the contract yet, so we're probably a little bit premature, but we are going to do this anyway, either way, shape or form.
Because I get asked this question a lot. “Do you have any good materials that you can point me towards?” And the only thing I point people towards is either my blog post or the papers I mentioned before.
So I wish I had a book I could sort of point people towards. And this is probably the one that we are going to try to write, me and João together.
João: We don't know the title yet, but it’s going to tell a story and bring these concepts together because the books and papers from academia go very deep, and sometimes this all can be a bit confusing, too much or out of focus for what we want to explore.
But we're going to bring these elements together and matching with the concepts that we talked about, such as the technical aspects of DDD, Team Topologies and Wardley Mapping.
And that's why we are doing the workshop because they will pair with each other (book and workshop). So actually doing the workshop, it's easier and serves as a foundation for the book.
Also, just to reassure all the social scientists out there that might be listening, we are not going to reinvent the wheel, of course. We are just putting it in the IT context by making it accessible to people. And all the resources are definitely going to be referred back to the paper. So it will probably be a long list of references in the book.
Trond's LinkedIn profile, where you can find some blog posts (as mentioned above).
Find some of Trond Hjorteland's talks on YouTube.
Learn with João Rosa and Trond Hjorteland
João and Trond are the trainers of the Intentional Architecture Workshop (Milan, 19-20 September 2023).
Check out the full list of our upcoming training courses: Avanscoperta Workshops.
Subscribe to our Newsletter 📩 (available in Italian and English) and be the first to know when we schedule a new workshop or we publish a new workshop.