BT

Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ

Topics

Choose your language

InfoQ Homepage Podcasts InfoQ Culture & Methods Trends in 2023

InfoQ Culture & Methods Trends in 2023

Bookmarks

In this podcast the whole Culture & Methods editorial team plus Rebecca Parsons, CTO of Thoughtworks, explore the trends they see happening in the technology industry in 2023

Key Takeaways

  •  Treating people well is not just the right thing to do, it has financial benefits as well.  Sadly this has been absent in much of the reporting about the layoffs in the tech industry lately
  • Generative AI tools like Co-pilot or ChatGPT work best if you know the answer that you want, they don’t know what they don’t know and hat makes using them risky
  • Responsible tech is more than just regulatory compliance and companies are having to be more socially responsible in order to attract and retain customers and employees
  • Project to product thinking is gaining traction
  • Hybrid and remote work are now the de facto standard, tools need to get better to support distributed work more effectively
  • Identify the kinds of activities that can effectively be done asynchronously, those that should be synchronous, but can still be remote first and those that you really have to not only do synchronously, but you've got to be co-located, because the exchange needs to be that rich
  • The article Culture & Methods Trends in 2023 is also available

Transcript

Shane Hastie: Good day folks. This is the whole InfoQ culture team and special guest for recording our podcast for the Trend Report for 2023. I'm Shane Hastie, I'm the lead editor for Culture & Methods on InfoQ. I'm going to shepherd our conversation today, but we'll start and I'll go around our virtual room and ask people to introduce themselves. We'll start with our special guest, Rebecca Parsons.

Rebecca, thank you so much for joining us. Would you mind just telling us very briefly who's Rebecca?

Introductions [00:36]

Rebecca Parsons: Well. Thank you Shane for having me. As you said, my name is Rebecca Parsons. I'm the chief technology officer for Thoughtworks and we're a software consulting company that we basically write software for other people.

Shane Hastie: Welcome.

Rebecca Parsons: Thank you.

Shane Hastie: Craig Smith.

Craig Smith: I'm the business agility practice lead for SoftEd, and I work with you as well Shane. Glad to be a part of the team.

Shane Hastie: And Raf Gemmail.

Raf Gemmail: Hi. Yes, I'm Raf. I am an engineering manager with Marigold in the MarTech space and many years technical coaching, Agile teams.

Shane Hastie: Thanks Raf. Last but certainly not least, Shaaron.

Shaaron A Alvares: Thank you Shane for having me. Very excited to be here and to be with all of you. Shaaron Alveres, I'm located in Seattle, Washington and I am director of GT Agile Transformation and Delivery at Salesforce. I'm working closely with executives to help them get better. I'm supporting of them and their teams in operation and engineering at GT.

Shane Hastie: Welcome folks. This is our motley crew of editors, reporters, and contributors for the InfoQ Culture and Method Space. It's wonderful to get us all together in one virtual space for a change. It doesn't happen often, but what's happening with the Culture and Methods trends for 2023? We were having a bit of a chat earlier and I think the big elephant in the room is the economy and the layoffs and everything that we've seen in that space.

What do people think? Shaaron, can we start with you?

The economy and layoffs in tech companies [02:11]

Shaaron A Alvares: Yeah, I'm happy to kick us off. You're right, Shane, with this new recession, there's a lot of changes happening in organization, priorities are shifting. Executives are more concerned at the moment about financial results, which led to a number of layoffs globally.

We've seen a lot of layoffs globally, in the US as well. Then that led to prioritizing high performance team, productivity, and efficiency. A huge impact I want to say on the culture aspect of organizations, on delivery as well, but a huge impact on culture and maybe a lost understanding how psychological safety is important for organization and teams.

Shane Hastie: Well, we've just come through two years where we've been focused on wellness and suddenly we've shifted. What's happened?

Rebecca, what have you seen?

Rebecca Parsons: Well, I think when the boom was happening, the tech talent was very clearly in the driver's seat. You add that to the pressures coming from the war in Ukraine, the global pandemic, increased visibility of these extreme weather events, the earthquake. All of these things have been contributing to a tremendous amount of stress. I think the response from many of the employers, both of the technology companies themselves, but also the enterprises who are trying to build up their own technology competency is they knew they had to fight to get the talent that they wanted.

Thinking about people-care was an opportunity to appeal to the needs of some of the people that they were trying to hire. I think as the macro economic pressures have changed, and we've heard this from some of the enterprises that we talk to that are not digital natives or not the technology companies, I don't have to fight with Facebook anymore for talent, because Facebook's just laid off all of these people or insert name of technology company here.

I think one of the interesting things that is particularly relevant from a culture and methods perspective is, okay, yes, you've got these people who are on the market, but are you going to actually be able to hire them as an enterprise? Even if you can hire them, are you going to be able to retain them when the market does turn around?

I'm reminded of a talk that Adrian Cockcroft gave several years ago where somebody from a more traditional enterprise said, "Well, yes, you can do all these things because you've got these wonderful people. And it's like, but I hired those people from you. I just gave them an environment within which they could be productive."

I do think we are going to see much more of a focus on how do we actually create an environment where we can get efficiencies and effectiveness, but we don't impact time to market, because I don't think we're going to go back to that era around the turn of the century, where so many places still considered IT as a cost center and that meant you had to stabilize and you had to standardize and you had to lock things down. Consumers are not going to accept that, even in this downturn situation.

Shane Hastie: Thank you. Craig, what are your thoughts on this?

Craig Smith: Firstly, to anyone's whose listening who's been through that, I think we share their pain and things that they're going through, but I think what I'm seeing from the outside is this is a little bit for a lot of organizations actually are a right-sizing that they needed to do that actually there's a lot of waste that we see in organizations. Whilst all those people were effectively working on things, the size of the organization probably couldn't sustain that.

What worries me though is that in this, were the leaders responsible for doing this, did they actually take the people from the right places? In other words, rather than just chopping off a whole side of a place here and moving it along, what I'm seeing in some of the organizations that I've been working with is that they'll just say, "Look, we're just going to get rid of these six teams over here," but they weren't the right places to get rid of.

The actual waste was in the actual mechanics of the organization. I think there's also going to be a realization of that moving through, but you can argue for a lot of those tech companies, particularly in Silicon Valley and the like, maybe they were just too big for what they needed. I think the way we're seeing some of these things play out in the media, whilst the way ... as Ben was talking about ... the way that's played out and the real culture of the company may start to shine through, it's also going to be interesting to see how this actually plays out in the end.

Do some of those organizations and now there's a much more radicalized size they have, was that actually what they should have been in the first place? Was there just too many people doing too many things for the organization anyway?

Shane Hastie: Raf, your thoughts.

Raf Gemmail: Listening to them about the culture coming out and the discrepancy in styles of letting people go. One thing I've been reflecting on is I remember back to mid 2000s, I was in the environment of US banks and there would be culls every now and then, where people would just pop, pop, pop away from their desks. We went through a transformation after that I saw, where there was a lot more humanism in the workplace. Looking at what's happened externally and to the people I know, the change management in letting people go in a more respectful fashion for some reason seems to be disappearing.

Yes, there may be an element of right sizing in all of this, but I completely agree with Craig, because when you are doing things with such haste, and maybe I'm projecting, inferring here, but when you're doing things at such haste, the knock on impact, which I have seen, is that you are left with teams which may not reflect the capabilities and the topologies and communication channels you'd intentionally try to create in the first place. You will have a consequence of that.

I'm looking at it from the perspective of we've been through one big shift with COVID and maybe it took us in a particular direction, which I think was potentially positive. Now there's this fear which is dollar driven, this financial fear, realizable fear. Maybe to some extent that is compromising a humanism as organizations, which have been trying to become more humanistic generally. That disturbs me a little bit.

Shane Hastie: Rebecca, can I put you on the spot? In your role at Thoughtworks, I'm assuming that you've had these conversations internally, have we got the right people and so forth? How do you keep that humanistic perspective at the fore?

Keeping workplaces humanistic when faced with pressure [08:29]

Rebecca Parsons: Well, I definitely agree it is one of those fundamental cultural aspects of companies. From the beginning, we have looked at our people as the most important asset that we have. If you truly believe that, and a lot of companies say it, but if you truly believe that, then you know that you have to look at things like capability and capacity and policies from that more humanistic perspective. I often get into debates with people when they're looking at diversity programs and they want to go immediately to the business case.

Yes, there's a very strong business case for this. There is a strong business case for having a strong employee value proposition, because turnover is expensive and replacing people is expensive. If you treat people well, you can make all of these wonderful business case arguments, but you can also look at it as it's just the right thing to do. If you truly believe that your people are your most valuable asset, what do you do with valuable assets?

You invest in maintaining them and nurturing them and helping them grow. You have to look at it from that humanistic perspective. Even now, particularly given the options that are available to technologists, treating somebody poorly in a layoff. It sounds like that is something that is happening in many of the organizations right now, based on the froth that you hear in the social media realm. You're just dooming yourself, in terms of your ability to hire, because there's only so much you can get back in terms of that credibility when push comes to shove, okay, we have to do a layoff. Then you've got all of these people going off on Twitter and LinkedIn about just how wrong and how cruel, needlessly cruel the process was.

Shane Hastie: An interesting article. It was a Wall Street Journal article I saw I think was talking about yes, there have been these massive layoffs, but also most of these people who are laid off are being re-employed very quickly. We're not seeing a lack of employment and a lack of demand. Is it just perhaps the people moving around. As Craig said, somewhat a bit of right-sizing? How do people feel about that?

Downsizing or rightsizing? [10:49]

Raf Gemmail: I think we're at the start of the curve. We've seen a big spike, lots of people in the market, but what I'm seeing is that trickling to other places. We're in New Zealand and I've recently heard of a local company in New Zealand, which is letting go of many people. It had grown rapidly to hundreds of people, same sort of deal, and people are being let go of.

These things often start somewhere and they spread. While they're starting to spread, they may be still later demand in the market. I'm hiring right now, but we are on that curve and so we'll see where we go. I think I'm confident that if a company manages and looks after its people well that you can create a good culture, but in the industry at large there are always waves. I think we're starting on that wave.

Shaaron Alveres: About this article. I didn't read it but I'm going to read it, so thank you for the reference,Shane, but what I think there are jobs, it's true that there's been an increase in job posting. I read that, but I think it's in small and medium size companies and mostly retail and service type of companies. I don't think it's in technology, software.

What I want to say is during the pandemic, those technology companies have hired massively, because we had to move to online, to the cloud, and there's been massive hires. Companies like Amazon, my company, Microsoft, they hired people they probably hired in two years close to between I think between 50,000 to a 100,000 people. Sometimes, in those large companies. Now, what had happened is there wasn't always a good onboarding process for those engineers. We focused on hiring and we expected them to deliver right away, but there wasn't a great onboarding plan, especially when we hire so many people and we set up brand new teams. I think that has created maybe a perceived lack of productivity and efficiencies.

Other companies, what I want to add is some companies chose to lay off massively, but very few executive chose to actually decrease their bonus or their compensation and not lay off those employees. They had a choice as well. We saw a lot of articles around executive who chose to not lay off their employees, because they had onboarding them. They already have acquired a lot of knowledge. That's good to see.

I'm trying to look at the silver lining around those layoffs. One thing that happened recently is I think companies organization, as they lay out their FY 24 plan, objectives, they started focusing a lot more on priorities. There's a ruthless prioritization that is happening actually within organizations to make sure that we focus on the right thing and all of our resources are really focusing on maybe doing less but focusing on the right thing. That's a silver lining. That's good.

Raf Gemmail: There's a counterpoint to focusing on the right thing. I came across a term recently ... I'm not prolific like the others, but ... the last piece I wrote was something a lady called Sandy Mamoli who was with us last year and she was quoting a guy from New Zealand who used the word "quarter fall". This idea of upfront planning, sometimes if those priorities are guiding where you go too stringently, that ability to discover, to create, to pivot may be compromised somewhat. I'm not always a fan of the ruthless prioritization sometimes, because I think innovation comes in the gaps in between.

Shaaron A Alvares: It has to be both prioritizing ... and when I talk about annual planning, I think most organization, I hope so, they're doing both. We set up goals for the year, but we review those goals on a quarterly basis. But I agree with you Raf, there's also pitfalls with less prioritization.

Shane Hastie: hink that gives us a nice segue to what could be our next topic. What is happening? What are the innovations that our listeners, our readers should be preparing for addressing?

One of the big ones I want to throw up there is ChatGPT. We can't have a conversation without somebody bringing in, "I type this into ChatGPT and here's what it says..." Just taking that as one example, what's happening?

The impact of ChatGPT and AI [14:57]

Craig Smith: Well, obviously the world of AI was always going to catch up with us in some way. I'm sure we've all read the articles about how ChatGPT is going to take away our roles and how, I think the one I just read the other day was that Agile coaches and scrum masters would be no longer needed, because you could just ask ChatGPT your questions. But I think going past just one particular tool, we're on the very start of the hockey stick in relation to AI and machine learning and all those types of things.

We've been collecting all this data, we talk about it, but now actually starting to turn it into tools like what we're starting to see, not only with that but also with image processing and all sorts of other things. How those tools can be used for useful purposes as opposed to, I guess some of the negative reporting we get now, which is how they use for evil, but also then trying to bring in things like copyright.

We read articles about how schools are banning tools like that, which worries me greatly because it's like when I went to school ... showing my age ... you couldn't use a calculator, you had to add up in your head. I can't remember how many times I've tried to do functions and equations in my head without a calculator or something in real life. Embrace the tools that help us do that, but don't copy the answers. Teach people about using these tools effectively.

It's going to be a real challenge across all of our industries about how we embrace these tools and use them to move forward as opposed to trying to block them or push them away.

Raf Gemmail: There's a security and privacy aspect to that as well, Craig, around what you're putting into these third party tools. I've been using it, experimenting is probably the word to use at this phase, but I had to give some feedback to someone who was interviewed for some other documentation. I've used the text side of it, which it's amazing at, but I'm very conscious with these things not to put real names in, not to disclose more than you would want to put into a Google search.

I was in a conversation yesterday which covered this sort of thing, that you don't want to put too much IP into these things. You want to have some guidelines. I think we're in that wild west where we're still trying to figure out how to use it, but there's definitely an element of how much do you put into those systems? How much do you edit it? I edit it after the event. I'm like, okay, this is what it said, now fit it to my context, because A, I don't talk like this. And B, I want this to be more personal and reflect the person I'm talking to, and what I know about them.

Craig Smith: While we were doing this, I just typed to the ChatGPT, "What are some of the innovations in culture and methods?" The answer was "agile culture, Lean Startup, design thinking, DevOps, holacracy and agile HR."

There you go. Conversation done.

Shane Hastie: Interestingly, that many of those are the trends that we have on our graph.

Rebecca, what are you seeing with ChatGPT and AI?

Rebecca Parsons: Well, we're doing a fair amount of experimentation both with the text capabilities of ChatGPT as well as the code-generation of both co-pilot and ChatGPT. The big concern and the big questions that we get and are uncovering ChatGPT in particular, is very confident of every answer it gives, whether it's utter nonsense or not. Sometimes, it really is absolutely spot on. Okay, yeah, be confident, you're right. Then it will confidently assert that the name of a country that is four letters that starts and ends with the same letter is "Chad".

Again, very confidently asserting, but it's utter nonsense. It does not know what it does not know. It knows areas it's not supposed to talk about, because it's got the guardrails in there, but it doesn't know what it doesn't know. That means from a code generation perspective, very often the mistakes are subtle. Occasionally, it will be blindingly obvious because it's generating some java code that won't compile or whatever, but very often it's things, off by one errors or something subtle like that, where if you don't actually know the result that you want, you won't actually know that the result it gave you is wrong.

Now, even with the level of quality that it has right now within a specific context, it can generate some really impressive code. One of our people in China put it through an exercise in typescript to basically generate a CRUD application, connecting to a Postgress database and oh, what do I have to do now? Now, how do I get this thing containerized? Took it through the step. One of the really powerful things about even the current instance is the ability for the model to get context, so that you're not always starting from scratch.

That means you can say, "Oh well that was close, but use this library instead," and it will keep the context of what it had before and take your new prompt into account. We're seeing people who are claiming though 60% increase in developer productivity. Well, I can imagine if you're doing something for which a canned algorithm exists out there on the net, that yeah you could probably generate it pretty easily, but to make a general claim about more broad software development activities, you have to take into account how much time do you actually spend typing versus thinking versus executing?

It's only going to be able basically to help you with the typing. Maybe if you're trying to learn a new language, an experienced programmer, we've got somebody who's actually trying to do that, trying to learn a new language by using ChatGPT. But again, it works best if you know the answer that you want.

Shane Hastie: What's happening in responsible tech? Rebecca, I know that you have a particular focus there.

Movements in responsible tech [20:46]

Rebecca Parsons: Well, people are starting to take it more seriously in the first place. They're starting to recognize the negative brand implications of breaches or other unmitigated consequences. Their investors are taking note of it, although there is some backlash in various places against ESG reporting, but investors are looking more broadly at how our companies are approaching their technology development.

If you leave hackers aside for the moment, none of us sets out to do bad things to people or communities. We don't do it intentionally. We're trying to do the right thing and sometimes the unintended consequences of what we do are good and there are additional benefits that we did not anticipate, but more often than not those unintended consequences are in fact bad. This whole responsible tech movement is around can't we as technologists be more intentional about trying to see what we wouldn't otherwise see?

The way I like to think about it, we are problem solvers as technologists. We see a problem, here's a stakeholder group, here's the pain point that they have, this is how I'm going to solve it. We have blinders on. Metaphorically, what we need to do is we need to lift our head up and look around the problem. Okay, what are the other stakeholders and communities that are being impacted by this and how? How do they relate to whoever my target audience is?

There are all kinds of wonderful facilitation techniques out there to help us as technologists see outside of our own head. One of my particular favorites is called the tarot cards of tech. They have tarot cards, but they're posing questions and posing scenarios that will prompt you to think about how someone else might react to your solution or what other impacts might these things have.

I'm quite encouraged at the fact that more organizations are taking this seriously, even if some of it does come from the perspective of compliance. We're seeing more and more regimes like GDPR, we have accessibility standards, and they're more or less enforced in different places. Some of it is coming from a compliance aspect, but some of that I think is also coming from this branding aspect. As more of the Gen Z's come into the talent market, they are expressing, "I don't want to work for a company that doesn't actually match my values and live out that particular value statement."

Responsible tech is the big part of that as well. So I am encouraged.

Shane Hastie: Craig, together we’ve been doing, but you've been very much doing a lot of work on ethics. How does that play in here?

Craig Smith: Yeah, I think a couple of things. Slightly segueing back to what Rebecca was talking about, one of the things I'm noticing a lot in organizations is that with all this compliance, a lot of organizations just have their head in the sand. It's just we have to maintain compliance and we particularly see this in regulated industries like finance and health, where when you talk to a lot of their technical teams, 90% sometimes what they work is just how do we keep the compliance in check?

I think the real challenges for the leadership of those organizations and the organizations in general is how do you then take those things and then innovate within them? Because if you stand back and go, well these particular laws around privacy and security or financial constraints, whatever it is, have been set up for a reason. Whether you agree or disagree, the ones that then innovate around that and go and think about how do I then satisfy the customer? Is the important part as opposed to just how do I satisfy the regulation that comes along? Work within the boundaries that you have being innovative.

In relation into ethical standards that you mentioned there before. For many of us on this call, Rebecca, you've been around since the start of this new work with us on the Agile Alliance, as well Shane and I. One of the challenges that you and I were seeing for many years is that there's been an influx of people who haven't gone on the same journey as many of us have, whether that be in technical or whether that be in agility. The older of us on the call here, we made the mistakes and righted them. As the industry starts to get older, we don't remember why we did certain things. The new programmers have never had to go through potentially writing things down to the machine language and understanding what happens at the base, because the compiler or the no-code just takes care of it for them.

In agility, they haven't been through the why did we go through a whole journey of things like story points, for example? It was because we were trying to solve a problem from way before, but now we've moved past things like story points. We're onto the next step, but they've missed the links about why we were there in the first place. When we start talking about ethical type things, it's how can we make sure that we still understand our base, that we still have this core skill set, but we are operating in an ethical way?

That's a challenge for lots of organizations, whether it be the work that now you and I have been doing Shane on coaching ethics, or indeed whether you're talking about programming, testing or any of those other parts of the puzzle that you might be filling.

Shane Hastie: What is the state of play in product management? We've had the project to product movement around for a while. We're certainly seeing it moving to the right on our trends graph, but what is the state of product management?

The state of product management – project to product gaining traction [26:09]

Sharon, maybe you've got some thoughts there.

Shaaron A Alvares: Yeah, I think it's still very slow. There's a lot of great practices that are getting traction. To draw on what you said about asking the team, I think that's not happening enough. That's still not happening downstream. It's happening upstream, when the product management hand off the requirements, but we still see too much of a silo between product management and then the product owner and the team. Not enough involvement of UX, developers, equality downstream actually. That's the key challenge I'm seeing that's still happening now.

There's a lot of great conversation happening though around discovery. I forgot how it's called, the three amigos, so it's getting a little bit of traction but still not enough.

Shane Hastie: Rebecca, what are you seeing in the product management and project to product thinking?

Rebecca Parsons: We're actually seeing it get more traction definitely. It has not crossed the chasm yet, but people are seeing the value in taking that more user-centered approach and thinking about it more from the perspective of how can I continue to deliver value? As opposed to can I meet this project plan and come in "on time and on budget"?

In fact, why are we building this? If you can't really answer the question of what the value is of this thing, it doesn't matter if it was on the project plan, we shouldn't be building it, if we don't know what value we're going to get out of it. I think it's as much as anything this focus on, we are looking at something that is a product and we are wanting to see how we can increase the value that our end users get from our product.

Value focus is causing this re-evaluation of how do I look at this effort that is now going to go in to create this product? That if I consider it a project, it has an end and it has a scope, whereas a product doesn't in the same way have that defined scope. It evolves as the value needs of the users evolve or as new users are brought into the fold. I think that in many ways the value mindset is part of what is driving, okay, I need to think about what I am delivering differently. This is easier to talk about in the context of a product life cycle than it is in the context of a project life cycle.

Shaaron A Alvares: In fact, we are seeing a more interest for a value stream. I think organizations are looking more at value stream. I think it ties into the mindset shift between project management, waterfall, and product management. It's great to see that.

Shane Hastie: That touches on team topologies, team patterns. Raf, I know that this is an area you're interested in.

Team topologies in rapidly changing environments [29:07]

Raf Gemmail: Going over the topics that we've just covered. We're in this environment where there is this financial pressure, there's this shift in what teams look like, teams becoming smaller, organizations restructuring on the fly. I think there's been some intentionality over the last few years to try and set up orgs, where the communications structures enable teams to focus on their value lines.

The fear I have right now is we're getting into a state where potentially the investment that has been there to try and create longer term efficiencies to enable teams to focus on that product may be compromised by how we restructure organizations. Another thing in there, just commenting on a point that Shaaron made there, was around value streams. Of the last year, I've been in a startup, I'm in a very large enterprise now. I could have been  right in front of your customer, which I love. Then the customer goes through proxies before the team sees them.

I've been trying to get a team to get closer to their customers and it's really effective, but I found the value streams really helpful both in both contexts actually. To actually draw to teams here, this is why we're working on this, this is the why. It trickles down to us. When we try and ship software and we continue to monitor in production, this is why it matters and how it feeds back into the loop. Even though we don't have that closeness that you may have in a startup, you still can try and build some of that empathy. I think that's really important.

But again, going back to those earlier points, I don't have that challenge, but I've heard of other organizations condensing and bringing teams together as they make mass redundancies. I do wonder what's going to happen with the team that has way too much cognitive load, that ends up having to manage too many systems. Yeah, we were nicely structured before, we had ownership of these microservices. Now, we're coming together and we need to know about 20 different things.

I have no data points, so I'll probably stop there, but there is a level of apprehension as we go through change. How do we preserve these structures that we've intentionally put in, to enable us to have good architectures, to have happy teams, to deliver value?

Shane Hastie: Rebecca, reconciling architecture with teams was a point that you raised before we started recording. Do you want to explore that with us a little bit?

Reconciling architecture with teams [31:13]

Rebecca Parsons: When you look at the way, particularly in large enterprises, enterprise architects used to work, it was architecture by decree. We have a very small group that was viewed as overhead and so we had to keep them as small as possible, because that's what you do with overhead, you try to shrink it and they rule by decree. Part of what I started to try to do, I had a hypothesis actually set many years ago that turned out to be wrong.

My initial hypothesis was that enterprise architecture groups in my experience at that time were generally ineffective. My hypothesis for why was that they didn't really have a shared understanding about what success looked like. I was giving a workshop at an international software conference, not an Agile conference. This was definitely not an Agile-friendly crowd. There were 53 enterprise architects from around the world. I asked them if they thought they understood what being successful looked like and that that was a shared understanding with everybody.

Everybody put their hand up. It's like, okay, well that shoots down that hypothesis. I didn't even get one person to agree with me, but then I asked the follow on question is how many of them felt they were successful? Three people put their hand up. These are clearly not people who were failing at their job, because their companies paid to send them to an international conference. Yet, only three people felt like they were being successful, which told me that the fundamental working model was wrong.

What I started to try to pick apart is what are the things that Agile teams can learn from architects and what are the things that architects need to learn from the teams? I think one of the most important things that the teams need to learn from the architects are what are the things that they're actually worrying about? Because a software delivery team has one objective. Get business functionality out as quickly and efficiently as possible to delight the users of that software. That's what they're there for.

The enterprise architects are there for protecting the overall long-term asset value of the IT estate. Completely different objectives, so it's not at all surprising that they don't often see eye to eye. Trying to bridge that gap. One important thing ... this was pre-DevOps movement, so this would be a little easier now ... but very often the teams didn't care if their log messages were effective. They didn't care how easy it was to recover a system or troubleshoot a production problem. That was somebody else's job. They didn't know what the enterprise architect was worried about.

Similarly, the enterprise architects, because they're so vastly outnumbered, because they're seen as overhead and therefore must be squeezed, they really didn't understand the reality on the ground. They didn't know how to get their concerns inserted into that process. That was really where that work started from is how can we get awareness in both directions? What are they actually worrying about and why?

Because at the end of the day, the architects shouldn't win all the time, but then the development team shouldn't win all the time either. But you can't have a conversation, you can't have that trade off conversation unless there's shared context. There was not that shared context. That was really the first step.

Shane Hastie: How do we maintain and increase that shared context?

Rebecca Parsons: To me, a lot of it is you have to get the enterprise architects in a position where they are seen as value generating, instead of cost savings, because as soon as you are associated with top line value, you don't have the same resource squeeze that you do if you're considered a utility function. Once you can start to get that, you can begin to have the levels of conversation.

I also think by the way, it takes almost a shadow architecture organization, because in a large enterprise with 50,000 developers, you're not going to be able to have a 1 architect to 10 or 20 developer ratio. There's still has to be other scaling mechanisms in there, but to me the first thing is to get organizational recognition of the value that you can derive from an enterprise architecture.

In many cases, that means being able to articulate what are the outcomes that I'm trying to achieve with the architectural decisions that I am making? If you focus those things on outcomes that are readily understandable by “the business”, then they are going to start to recognize the value.

In remote environments process and documentation matter more [35:53<]

Raf Gemmail: I'm keen to share something I've seen recently in that space, Rebecca, which is I've worked in enterprises which had the enterprise architecture function and seen that mismatch between teams focused on the now and being the person between trying to shepherd them towards the architecture. I'm a big believer in your work and the notion of the emergent architectures, but you've got that contention. I'm in an organization now where there's a really lovely distributed function around democratized function, around looking at that broader architecture.

We're a family of products which is global. We're fully remote. This is quite impressive. It's works and I think it's works partly because it's remote, but we move that thinking into more asynchronous processes, RFC processes which are proven and defined most of our standards. Collaborative processes, people providing feedback. Process is probably the word here. One thing I've become comfortable with is we've always said, "It's just over," when people interactions over processes and tools. We said "It's over," it's still important, but there was a bias towards the people interactions.

What I've found since I've gone remote is that as we set up processes, we set up documentation, people work at their own pace, you set the boundaries that you don't need to always consciously shepherd. Then people innovate within those boundaries. What I'm seeing, are great architectural decisions, evolution of historic architecture, really powerful conversations that allow teams to get involved, as well as senior engineers resulting in technical outcomes that actually hit backlogs.

It's an engineering focused culture with a good focus on trying to deliver for product, but I've not seen that to this day. It's working really, really well. I think there may be more evolution to things becoming more asynchronous and slow. I've been looking at GitLab and GitLab is very focused around documentation driven cultures, allowing this async to happen. If remote continues the way it has, maybe we'll move into some of these functions, which are seen as overhead, perhaps being more democratized and shared across teams and across the stakeholders are actually impacted.

Shane Hastie: That's a really nice segue into what is happening in terms of hybrid, remote, in person. We're seeing the mix and certainly one of the trends we've been looking at is that shift to more asynchronous work.

What do people feel about that?

Hybrid & remote work is now the de facto standard, and it’s still hard [38:12]

Craig Smith: It is now just the de facto standard. I don't want to say "new normal," but the thought where everyone throws around. We were moving that way I think the last couple of years anyway, it's just moved it forward. Again now, it comes back to a segue and a call back to some of the things you talked about earlier, is it starts to show the culture of organizations as to whether they force people back into the office or whether people come together, because it's the right thing.

Now there are some things, I'm sure we all know where talking face-to-face really does make a difference when you're really trying to make tough decisions or trying to do things quickly. Yet we didn't have to get on an airplane to do a recording like this, which even maybe five or six years ago, would've been done around a microphone and not in Tools like the ones that we're using today.

My biggest concern that I'm seeing in my role, though, is that organizations still haven't truly made this hybrid working option as good as it can be. What worries me is that if we can't do it through one vendor, and I'll call Microsoft out here, that if you can't do it in Teams/Whiteboard/SharePoint, then you can't do it. There is not one vendor yet that actually has the solution nailed, whether in that market domain or another one.

Organizations still, I believe haven't all made the switch to tools that really help people work remotely. Rebecca, you were talking about the tools like co-pilot and other things that help engineers. The reality is that once you get outside of the true tech companies, that the people who are just trying to sling code for your bank or insurance company or a small to medium size organization often don't have the luxuries of those tools, whatever they might be. They're trying to pull things together.

As a result, they're the people who are also riding the code that drives our cars and runs our traffic lights and all those kinds of things, which does worry me. Investing in those and making the right decisions is one of the big challenges, whilst also laying on top of that all the challenges that we have in technology, like cybersecurity and other things that go along with it.

You don't want to have too many tools. The argument that tools like say Miro or Mural, those type of sharing tools is that most organizations don't want to turn those on, because you can upload a file and drop it onto a board. When you block those things out, it makes those tools less usable, because I need to share an image with Shane in order to make this thing happen.

We've still got challenges around how we do this. Again, it comes back a bit to the ethical thing, trusting people to do the work whilst at the same time making sure that we don't give away all the corporate secrets and throw them into it all the tools.

Shane Hastie: Sharon, you're in a mega organization that is tackling all of these things. What are you seeing?

Shaaron Alveres: Yeah, speaking about my organization, I want to say that surprisingly we introduced a lot of great practices right at the beginning of the pandemic. We design and rolled out a program that's called Success from Anywhere. We revised the program, we talked about reviewing priorities, a plan and so on. We called it a "connect from anywhere" when we were in the middle of the pandemic because we realized that working remotely introduced a lot of disconnect between people.

We moved from "success from anywhere" to "connect from anywhere". We developed a lot of great content to help teams stay connected and also new hires onboard and feel connected with their teams but also with their managers. We also focused a lot on the team working agreement. I would say it's a small activity, but in this scheme of software development, it's just a team working agreement, but we realize how important that activity is anytime, but especially when we go through a pandemic. How do we want to connect? How do we want to work together? Because we are completely remote, so we started being 100% remote, but then progressively we introduce some hybrid ways of working. How do we transition from a remote to hybrid? How will we ensure that our teammates who are located, who are not in the US or who are remote still feel connected, included?

I have to say that we've done a lot of great work actually. Now, I see a lot of organizations moving to hybrid or 100% back to the office. It's interesting, because I was making the comparison between what we saw ... the bad hybrid and bad return to the office, was which led the great resignation. Now, we are talking about the great layoffs. It's interesting how events leads to reactions from people, organizations.

Ways of working hybrid and remote is a really, really important topic to solve. I hope that companies are going to keep investing in that because it's really, really important for the health of the team, for their ability to work together, connect, but also innovate.

Tools can help, but they’re not a panacea [42:52]

Raf Gemmail: I might share from our perspective. I'm in a globally remote team as I said, but my local team, my immediate peers, my first team, based in the same time zone. I'm hiring now and an interesting consideration as I do this is we do async fairly well, but does one open up the local team? That small team trying to keep within Dunbar's number, does that one keep that communication loop in the same time zone or does one allow that to seep into other time zones?

For now, the focus is teams which are localized, collaborating asynchronously, interfacing each other. I don't know how that may work for others. I've heard of other teams trying to do global remote, but I think you end up with a lot of autonomous people. The collaboration may potentially be compromised, the real time collaboration. We are asynchronously able to communicate with others because ... I think we use Slack reasonably well. The things that you can do. I remember Manuel Pias did a talk from some years ago around being intentional about your topologies, your conventions in something like Slack, making the channels discoverable, making sure you didn't have too much noise.

We have fairly good access to people via Slack. We have the processes, as I said previously that allow us to function asynchronously. Our standups, we have standups currently, four standups which are on a call. One and it may be more later, which is fully async. The async one works really well. People continue with stuff, we talk, but when we have interaction points later in the day, you don't feel that sense of loss. Some of my teammates, I hadn't met for months after I joined this context. I felt like I knew them by the time we met in person.

I've always been like, "We should get together." I remember I went to an Alistair Cockburn talk years ago, where I had a distributed team and asked, "How do I get this team over here and this team over here to get together?" He used the word "trust budgets", for years, I parroted, "We need a trust budget." It was actually really valuable bringing people together. I still think it is, but our tools have evolved to the stage where it feels first class to have an interaction digitally and to be able to call it and stop.

Obviously, you've got the Zoom fatigue and you don't want to go over, but if you can keep those boundaries right, we can work really effectively remotely. Documentation, source code, pull requests, jumping on calls, swarming where we need to, mobbing where we need to, pairing where we need to, but being intentional and using the tools to our benefit rather than anything else.

Another point, just to Craig's thing earlier about using these tools, these third party tools, I see tools like Loom, which are really powerful and people use them to share content. Sometimes you're like, "Oh, has that got something sensitive in it?" It's a hashed URL, it's not within the org scope. I think one thing that needs to happen there is being intentional again about what tools that in that organizational context we're using and managing them intentionally, because security and privacy always is the afterthought. With these, if we're using them, let's sign up for licenses. Let's make them part of the org thing. Let's put SSO in front of it, let's stick guidance in front of those.

Shane Hastie: Rebecca, how is Thoughtworks handling this? You're massive, global and you're dealing with thousands of different customers as well.

There is no one-size-fits-all approach and you can’t get away from time zones [45:55]

Rebecca Parsons: Yeah, and it varies by country, it varies by client. In places like North America and Australia and Germany, our clients are not saying, "You need to start coming back into our offices five days a week." In fact, ironically we have more of our individual client people working in our offices, because we do have all of our offices open now. We are not requiring people to come into the office. Certain countries are trying to encourage that. Other countries have said, "We're going to accept that we have people who will always be fully remote."

We have, what I would say is lots of different experiments going on across the company on how to do things. One thing, and this is just probably because I spend so much of my time working with global groups, I'm exposed to it more. The one thing that you can't get away from is time zones. That's why I think this emphasis on can we look at some of these processes and make them more asynchronous is so important. But for those things that really do require that co-creation that was spoken of earlier, you can't get away from time zones.

I've had people say to me, "There will never again be a global meeting." And I said, "Well, actually it's dangerous to use the 'N word' anytime. But no, you can't get away from time zones." There are going to be things. I think a lot of what we're going through at the moment is identifying these are the kinds of activities that we can effectively do asynchronously. These are the kinds of activities that should be synchronous, but can still be remote first. These are things that you really have to not only do synchronously, but you've got to be co-located, because the exchange needs to be that rich. I think we're looking at many of our different activities and this is what it takes to do this and this is what it takes to do that.

One final thing that I found interesting, we did experiments again across our development community on remote pairing tools and there wasn't a winner. Different people would swear, "This is the most wonderful tool available." Another group would swear, "No, this is the most wonderful tool available and that thing is horrible."

I think one of the things that we are going to also have to grapple with is, it's one thing to tell somebody, "Okay, I'm pairing with you Shane, and we're sitting next to each other and we need to be able to swap keyboards." There's a good compelling reason to make sure there's standardization. As we get further and further away, the needs for, okay, everybody needs to use the same too,. well, the computation is different in that context. I think we're going to be seeing a lot of innovation in our ways of working as we sort out how to do these different activities.

Raf Gemmail: I'm hoping Rebecca, at some stage I get to pair with people in virtual reality. I've got a way of exporting monitors into here, a virtual setup. I can be floating in space and working. I did a demo for some people at work, but I think we're still in that stage, where there's a lot of skepticism. And Meta gets a lot of bad rep, because the images that are shown look rudimentary, but I've tried out their Work Spaces product. When you're in, it's a collaborative thing. The whiteboards, they're other things I've used, but I'm still hopeful we'll get to the stage where some of the collaboration is a bit more immersive.

Rebecca Parsons: But you still can't get away from time zones.

Raf Gemmail: You can't get away from time zones. Oh, one more thing to thread though is that we talked about trust in the team's building cohesion. As a remote person, it's really easy to feel or to be worried that you're going to feel excluded, but we're doing things like painting events remotely. Sibling team did a cooking event. The games and all sorts of things which have really, I'd say for people are trying to do the remote, make sure you do the social aspects as well, because it builds cohesion, it's fun. I've received in the post canvases and paints.

Awesome! I can't paint, but by the end of it, I had something and I did it with my team.

Shane Hastie: We have come close to the end of our time box, so we'll do a quick whip around. Hopes and wishes for 2023 and beyond. Shaaron, can I start with you?

Hopes and wishes for 2023 and beyond [49:59]

Shaaron A Alvares: I really hope if I had a magic wand, I would want all organization to support our teams, our Agile projects, our development teams through the recession, to the priority of productivity and high performance. I think it's not just about productivity and high performance. I hope we're not going to stop looking at wellness, mental strength, and also psychological safety. That would be my wish.

Shane Hastie: Raf, your turn.

Raf Gemmail: My hope for this year is that, to echo Shaaron, that we see a rebalancing in how we work. After this right sizing that's going on in the industry. It settles, my hope is that humanism is still valued and that we work together and we value working together as people. The dollar target will always be there, but how we create the teams that are going to achieve that and deliver to our org and stay within our org, remains the focus.

My hope is that we remain responsible in how we use tools. We were talking about ChatGPT, and it's going to come regardless. I'm hoping as we embrace this new technology, we're responsible as to how we use it, how it impacts our teams, how it impacts our planning. I'm hoping we see innovation beyond looking to a few solutions that have come from a few models and we see innovation from other companies trying to localize and train and create models from corpus' of data for their organizations and context.

I'm hopeful we get to the end of the year after a few rough years (we just had a cyclone in New Zealand) which I'm getting out of and we see a pathway to 2024, which is good for the whole planet actually. A good thing I've shared to finish is I saw recently a friend of mine got a technical certification. Now, there's so many of them out there, but it was a technical certification around green ethical development. I thought that was brilliant to see. I'm hoping we have that more as a focus going forward.

Shane Hastie: Craig.

Craig Smith: Well, Shane, I asked ChatGPT what my hopes are for Culture and Methods were in 2023. It told me that as an AI language model, it doesn't have access to my personal hopes and aspirations. That's a good start, but what I hope not to see is that as we have these challenges Ben was just talking about and then everyone was just talking about, is that we resist the pull of the past. That there's been a lot of good work over the last few years for individuals and organizations to embrace new ways of working and moving technology forward. When things start to become unstable, we tend to often batten down the hatches. I'm seeing that in organizations that we talked about product before. Yeah, it's like, no, let's tighten up purse strings and now do a lot of upfront analysis and go back to old project methods.

You could use a lot of different things like that, the way we structure teams, because we're talking about all the layoffs and things is just let's just chop off the left arm and remove innovation and all those things. I think the individuals and teams that are going to win are the ones that continue to push forward and continue to innovate. That's my hope for our listeners, as we move forward is that we still have to chip away the basics.

When you work with organizations like Thoughtworks and things that you can be lulled sometimes in a false sense that all developers are great and they're all using lots of tools and that DevOps is ruling the world. That isn't in a lot of places, but when you actually get to the rats and mice of things like government and large corporates, we're still doing things like we were doing them 40 years ago. We're making slow inroads, but we still have to chip away at all of those basics, because those organizations have to keep at the same pace as everybody else. If not, we all get dragged backwards. I hope that the people in organizations continue to innovate.

Shane Hastie: And Rebecca.

Rebecca Parsons: A couple of unrelated things. One thing I hope for is that we can continue the move towards more user-centered development, that we can continue to focus on what is the value that we're actually delivering? Use that to guide our decision making, as opposed to re-trenching too far into, oh, I've got to go back into cost cutting. I've got to go back into lockdown of my systems, because expense. I think there are some forces in the world that are pushing that direction, because the changes to those expectations of our users are not going to slow down.

Organizations are going to have to continue to respond. I have some hope on that. A second, although this is a pipe dream, but I really wish that our industry could get away from the belief in the one true way or the silver bullet, or, oh yes, if I apply this, a miracle will occur. We still, as an industry have not accepted that there is a lot of actual complexity in the systems that we are building. Yes, we have accidental complexity and we want to continue to chip away at that, but there is essential complexity in there as well. We have to accept that. There isn't going to be this magic package that you can throw at a complex problem and have it all go away.

Finally, I hope we can continue to raise awareness across the industry of the importance of, in particular, security and privacy. Because there's still so much low hanging fruit out there for the hackers to exploit and we're all human beings. We are going to make mistakes, but if we can continue to raise the awareness, find new tools to automate, particular kinds of scans, to reduce exposure, whatever we have to do, but as an industry, I believe we need to take that more seriously. I hope we continue to advance on that journey.

Shane Hastie: For myself, I would echo pretty much everything that you have said and add the wonderful opportunity that it is to work with such a great group of people in the culture and method space. Thank you all so much for the contributions over the last many years.

Rebecca, thank you for joining us as our special guest today.

Rebecca Parsons: Thank you for having me.

EDITORIAL NOTE:

When we recorded this podcast we were also joined by Ben Linders, however a technical glitch meant the file with his voice recording did not get saved so we lost his contribution.  Watch for a future podcast where Ben and Shane discuss their views on the current and future state of culture and methods.

About the Authors

More about our podcasts

You can keep up-to-date with the podcasts via our RSS Feed, and they are available via SoundCloud, Apple Podcasts, Spotify, Overcast and the Google Podcast. From this page you also have access to our recorded show notes. They all have clickable links that will take you directly to that part of the audio.

Previous podcasts

Rate this Article

Adoption
Style

Hello stranger!

You need to Register an InfoQ account or or login to post comments. But there's so much more behind being registered.

Get the most out of the InfoQ experience.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Community comments

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

BT