Brief summary
It's often said that technology moves quickly, but the decisions we make about it can have long-term consequences. That's why identifying these trends — and understanding what they mean — matters. At Thoughtworks we do that with our Looking Glass report. Our most recent edition was published at the end of 2022; it provides a useful framework for thinking through the major shifts that look set to happen across the industry in 2023.
In this episode of the Technology Podcast, regular hosts Mike Mason and Ken Mugrage take the guest seats to talk to Neal Ford about the six key trends in the latest Looking Glass. They explain why they were chosen before diving deeper into a selection of some of the most hotly debated trends that feature in the report, such as metaverse and Web3.
- Explore the latest Looking Glass
- Read Mike Mason's metaverse article (co-written with Barton Friedland)
Episode transcript
[Music]
Neal Ford: Hello, and welcome to the Thoughtworks Technology Podcast. I am one of your regular hosts, Neal Ford. You're going to hear a lot of familiar voices today because we're swapping hats and two of your regular hosts are sitting in the much more comfortable cushier guest chairs today in our luxurious podcast studio, and they're going to be here talking about this document called Looking Glass. I'm joined today by Ken and Mike. Hi, Ken, Mike.
Ken: Hello.
Mike: Hello. Hello.
Neal: They are regular hosts, but today one of their mini-hats within Thoughtworks is they are members of OCTO, which is the Office of the Chief Technology Officer. One of the things that OCTO is tasked to do as advisory to Rebecca, our CTO, is try to imperfectly look into the future as much as possible to help us make decisions about technology — where should we be going and where should we be putting our effort, et cetera. That is an internal document, publication I guess we could call it, called Looking Glass. In the grand tradition of Thoughtworks, things that are valuable we like to make public because we believe a rising tide raises all ships.
We started publishing Looking Glass to give you some insight into our decision process, and that's what we're going to be talking about today with Ken and Mike, is the most recent edition of Looking Glass and this idea of “lenses”. There are six lenses in our Looking Glass, each of those is a perspective on technology in the near-term future. We're going to talk about the six that exist on our Looking Glass, and then we're going to do a little bit of a deeper dive into a few of them. Let's talk about the lenses that show up on this most recent edition of Looking Glass.
If you want to read the actual document it is in the resources link for our podcast. Let's talk about the first one here, Platforms as Products.
Ken: Sure. I'll take that one. This is Ken. It's funny because some of these repeat, which is, we think, a good thing. I should say, right up front, if you read last year's Looking Glass, some of these may look familiar. We actually say internally that if they don't repeat a little bit from year-to-year, we probably were wrong. While there is some short-term stuff here, there's hopefully some longer term. There is a little bit of a different focus — this year with Platforms as Products, we're focusing on them as products, which means treat the people that are using the platform as real customers, get constant feedback, do internal training, do internal marketing. Really a product-focus on developing and using platforms.
Mike: I think that's important because lots of folks are investing a lot in platform builds and platform rebuilds and often are disappointed with the outcomes. This is something that Thoughtworks talks a lot about both on our website and on podcasts and in this report, is how to do platform-building well. This year's advice is to have this product focus to think about the customers for it and so on. Previously, we've had different angles of advice around aligning on what you want to get out of your platform and that kind of thing, but this year's is all about that product focus.
Neal: I think as much as anything, this is about putting out our perspective because I don't know of any other term that has hit so much semantic diffusion out in our world as “platform” now because everybody talks about platform as the end-all solution for everything. We talk about platforms too, but we have a very specific perspective on platforms and the way they're done well, the way they're not done well. I think that's what we're trying to convey here.
Mike: I almost think you shouldn't use the word platform on its own because it's not information-rich enough. Do you mean infrastructure platform, do you mean API platform, do you mean platform business, which is something entirely different and much less about technology and much more about business model? All of those are flavors of platforms. Very important to say what you mean in this space.
The next lens is Hostile Tech. That's something that has also been in the Looking Glass for the past three years, and that's really around tackling challenges that arise from security ethics and privacy, all of which are important in the technology space. I think just the phrase “hostile tech” itself raises some eyebrows, and that's deliberate on our part, and we want to make it clear that depending on who you are, you might consider some of the things running on your computer, running on your smartphone, some of the systems that you interact with on a day-to-day basis as actually being hostile to you depending on your perspective — because if a business is selling advertising and selling my eyeball time, if I am unclear about that or don't feel like I'm getting value for that, that's maybe a little bit of a hostile act.
That's why we use that name for this and dive in a little bit to some of the issues there. It's a very big space. We have to take a particular angle on that each year, but it is important to think about all the stuff around technology.
Ken: Mike makes a really good point there where some people's opinions will differ on these. Inside my own home, my partner very much values seeing advertisements that apply to her so she can get the discount on the thing, whereas I don't want to be tracked and to let them know what ad I saw on somewhere, some social app, so I can see the other one. The real key here is that, of course, things that you think of as illegal, hacking, malware, those sorts of things are hostile, of course, but we want to openly call out the misuse of private data as hostile.
The next one is “Partnering with AI.” Artificial intelligence is everywhere as we all know. The real call out to the lens this year, though, is that it's really becoming more mainstream and we think it needs to become or should become more mainstream in our clients and partners and so forth. What that means is we don't have a “department of artificial intelligence”. We don't have a group off in the corner somewhere that does the AI, or some mind over there, some electronic mind that's doing the AI… you really need to think about AI as part of your everyday projects. How can it help your people be more effective?
There's examples everywhere from clothing retailers that get suggestions based on your past purchases and then help you layer it, down to using AI to help write code which I think Mike's been playing around with a little bit lately.
Mike: Yes. I think we're going to do a deeper dive into this once we've finished going through the lenses here, but this one's super topical. Three of the lenses we've picked on as being newsworthy recently, and we'll dive deeper into them. We'll come back to this one in a minute. The next lens that we include in our looking glass is called “Making the Metaverse.”
Last year and still, now, there's tons of metaverse hype — people using that word to apply to an awful lot of things. Really, what we think is important is the digital interactions becoming deeper and richer. That's really the thing that is important here, rather than maybe just the splashy headlines around what people are doing in the metaverse.
I also think metaverse is one of these words that has gone in the platform route in which some people are… using “metaverse” to mean just almost anything to do with digital tech these days and some folks thinks it's all about the 3D embodied internet, which to me is the thing that I identify most closely with metaverse, but then there's also folks who want to tie the metaverse and Web3 together as concepts and make them reinforcing or even required things one to the other.
I think — and we'll dive into this in a bit more detail later — but this is another one of those ones where you really need to cut through the hype and demystify some of what's going on there.
Ken: Which I'll actually use as a bridge to introduce the next one because we are going to go into it a little bit deeper. The next one is “Evaluating Web3.” The key here is to understand that where Web2 was an observation by some folks on what they thought had happened to the Web, the things that had been added to e-commerce and other [things], and they're like, "This is a 2.0 of the Web." In this case, there are people saying what they think Web3 should be. It's people with their own biases and reasons for it, which we can talk about how justified and evil some of those are. No, I didn't say evil, but I did say evil. We think there is more to it.
If you think of Web3 as the distribution and the decentralization of the internet and decentralized identity and so forth, there is some value there. Trading crypto is not that value, if you ask me personally, but there is some value to some of the underlying technology.
Neal: That's the other one we're going to do a deeper dive into in just a moment. Let's wrap up the last one here, and we'll do a deeper dive into the three.
Mike: Sure. The last lens — and these are all kind of storylines within the technology landscape — is “Accelerating Sustainability.” That's both about [chuckles] the accelerating need to tackle the climate crisis and move to a more sustainable world, but also the accelerating response that we're seeing through technology, through policy, through the kinds of things that people are thinking about. Lots of organizations have a Chief Sustainability Officer these days.
Actually, if you look at the CES coverage from last week, actually, sustainability was a really big watchword there with lots of brands getting on board with that as being part of their positioning and their sustainability stance being something that they are pitching to consumers as being something that those folks who are choosing between brands and choosing providers and services should be thinking about sustainability. The Looking Glass lens there talks a lot about that and what's going on in that space.
Neal: I was going to say the important thing there, I think, is both in the large and the small because Amazon in particular has been focusing a lot on this. The choices that you make as a developer or an architect can have big implications as to how hot the data is and how cold it is, because the colder it is, the more sustainable it is because there's not on spinning platters. Some seemingly trivial decisions can have long-term implications on things like sustainability and it's something that awareness not just at the corporate enterprise architecture level, but at the individual decision level can have an impact.
Ken: Yes. One thing, to plug it — please go read the entire report! —is that we try to give action advice on all these things. It's the time of year when all these types of reports come out and got the Consumer Electronics Show and those kinds of things. A lot of that's observational. A lot of that is, "This is what happened last year." There's a lot of advice in here about what we think that clients should do and what readers should do and what we need to do internally.
On sustainability, it's funny that Neal brought up Amazon because I live in Seattle, which is the headquarters. Amazon recently sponsored one of the sports stadiums. And they didn't call it Amazon Stadium or something like that. They called it Climate Pledge Arena. One of the things under sustainability, the advice is there is a market value, there is a consumer awareness value to being sustainable over and above the amount of carbon you're using and everything else. It's funny that Amazon's got 1,000 trucks winding around here.
Mike: I was going to say this is a podcast so people can't see the look on Neal's face when you were mentioning Climate Pledge Arena, because the cognitive dissonance that was happening in his head was pretty serious there! [Laughter] That speaks to the branding importance of sustainability, if nothing else.
Ken: Exactly. Yes. That was the point is that for a lot of these lenses, when we're coming up with it, it's like, "That's not really metaverse, is it? No, but when they say it's metaverse, they're much more likely to get investors.” That is true. That is the business reality of many of these lenses. Some of these are about what you look like, perception is reality and that sort of thing.
Mike: That's a good segue back to-- Sorry, Neal, go ahead.
Neal: I was going to say exactly what you were saying, that's a good segue back to the metaverse, and we were talking about semantic diffusion earlier. It seems like in the metaverse they're almost encouraging semantic diffusion to happen as much as possible so that you can just label anything as metaverse and get away with it. I don't think I've ever seen a term move from meaningful to meaningless in such a short period of time.
Mike: I'd agree, and what we tried to do there — I wrote an article on this a couple of months ago, we tried to disambiguate the main usages that we've seen of the word metaverse. The first one, I'm just going to call it “Zuckerberg's embodied internet.” That's the avatars in your living room, holograms in your living room, 3D everything, Ready Player One. It's the meta announcement from last year when they did the name change from Facebook. It's all of that kind of stuff. It's the high-end Hollywood graphics version of the internet, the embodied internet. That is something that people often can be meaning when they're talking about the metaverse.
The second potential meaning of the metaverse is actually to do with Web3. That's the ownership economy, evolution of the internet, you need to be able to sell things to people online for them to bother to spend time in the metaverse, and all that stuff. There's the money version of a metaverse. Then the third major use of metaverse is actually the industrial metaverse. That's things like NVIDIA partnering with Siemens to build a virtual factory, and to be able to map that to a real physical factory, and then go back and forth between the physical and the digital and get value out of being able to do things within your digital version of that factory that can be used to then deploy to the real-world version of the factory.
The canonical example of that actually is Tesla, where they made a change to the way they were building the cars that rerouted some of the cabling, so they could take the car from 200 to 250, I don't know what the units are, megawatts sounds a bit too much. Anyway, they significantly increased the charge rate on the vehicle, kilowatts, that's it, 200, 250-kilowatt charge rate.
That was actually like a software change for all the robots. They deployed that software change to their continuous delivery pipeline, which you can think of in a way as an industrial metaverse simulation thing, which said, "Yep, everything will be fine." They deployed that to the real factory, and then three hours later, new cars were coming out that had a new characteristic based on this digital software change that they had made. I think that's the main thing with the metaverse, right: you’ve got to figure out what is somebody talking about — are they talking to you about the 3D thing? Are they talking about this Web3 thing? Are they talking about digital twins and industrial stuff?
Mike: It was funny because when we were trying to figure out the advice for this one, it's very, very situational. We say, "Okay, hey, if I'm a big shoe manufacturer, should I go open a store in the metaverse again?" Whatever that means. You know what, the answer might be yes. Meet your customers where they are, if the people who are collecting your brand of basketball player-branded shoes, not to mention any names, that's where they are because they're playing games there, and they've bought onto the Zuckerberg train. Then, by all means, set up a storefront there and play with them.
I was watching some of the stuff on YouTube today from CES. I don't know if anybody would call this metaverse, but one of the things we talked about is for most people the augmented reality part of the metaverse is probably the more impactful. In the caterpillar booth — Caterpillar had a booth at CES, so first get your head around that! Part of their booth was you could sit down in a control center, and you're driving a tractor that was in the next state over, it was in Arizona, 400 and some miles away.
Not autonomous — they've had autonomous dump trucks for quite some time, but this was an actual operator who could look around and see what they were doing and move Earth. Next to them was somebody doing the same thing with a backhoe that was in a different state. Is that metaverse? I don't know, but they're sitting there in an augmented reality type of situation, looking to their left, seeing what's on the left of the piece of machinery, and using levers and moving that rock. It's pretty interesting.
Neal: Well, that to me seems like-- the analogy that I go back to is I used to read a lot of science fiction when I was a kid that was written in 1950s. A lot of those were ships that were traveling through the galaxies, and all the scientists on the ships had slide rules with them because you could imagine the spaceship, but you couldn't imagine the pocket calculator.
The obvious, easy things are the ones that you miss and not the big grand things. So much of the metaverse stuff is the big grand — physical presence, going to a concert in the metaverse setting aside for a moment that sounds like a terrible, terrible way to say a concert, but okay, if the capability is there, some people might take advantage of it. I think and Ken touched on this, I think we're at least the early win is going to be around augmented reality and helping reality rather than replacing it.
I think that's also true with another one of our lenses about AI. Rather than having AI replace developers, it's going to become a really useful research tool for developers, sort of like Google on steroids. Of course, the obvious recent example of this one is ChatGPT and Copilot.
Mike: Yes. We've had an internal thread on this because as a software company, well, we've now got AI tools that can just write the code for us, so are we all out of the job here? I'm pretty sure most of the listeners would be skeptical that programmers are going to be out of a job having been told many times in the history of their careers that there's some new thing that has come along that will render programmers obsolete and we don't need them anymore. Hello Low-code, Hello 4GLs, whatever the-
Neal: COBOL. That was the original selling point of COBOL was get rid of the pesky programmers and let the business people write the code directly [Laughs].
Mike: Exactly. The reason that we caught the lens partnering with AI is because, and we've been using this word partnering for a while, is that the point is to let both humans and machines play to their strengths. The phenomenal thing that the machines can do is crunch tons and tons of data; they're really great at pattern matching. So when you look at something like GitHub Copilot, that I've been using pretty extensively recently, some of the stuff it can do is fantastic, but I would say it has about a 50% hit rate.
On being fantastic versus producing something that won't even compile or was completely wrong or is just made up or guessing doesn't fit my code base, all those kinds of things. Right? Even when these things-- so ChatGPT, for example, I was trying to get it to write some code for me and it did an okay job, but any programmer would be able to look at that and see all of the limitations in what it had produced.
It didn't answer the problem that I was really asking it from a simple one-sentence description of what I was trying to do. And so what I think is actually most likely to happen there is that there is a place for these tools in accelerating professional programmers in getting their craft done, but they are not things that you just let run off on their own. Funnily enough, the code that ChatGPT produces is exactly the same as the English language prose that it produces — which is plausible looking, highly self-confident, but wrong in subtle ways. It does that with English, it does that with program code too.
Ken: Yes. That's what I was gonna say. When it creates something that won't compile, okay, I know that's wrong. No problem. To me, and this is where a lot of the lenses play with each other because is there a hostile tech angle to AI-generated content and code, whatever? That's the problem is when it's plausible. Like if you use some of these tools like there's a ton of them out there marketed right now for people that are writing content.
Mike and I did a lot of the authoring on Looking Glass and we actually played with a few of them. We gave them bullet points and said, "What would it say?" The danger is that quite often it was plausible. If I didn't know anything about the topic and I read it, I was think, "Wow, okay, this is really insightful." The problem was it was wrong. They're only as good as the data that they find.
And so what is that AI trained on? Is it trained on a model that was good? By whose definition of good is also another question, of course. The danger to me isn't the stuff as I mentioned that that doesn't compile, the danger is the stuff to me where someone like me who hasn't written real code in many, many years will look at it and go, "Okay, that makes sense." I'll deploy it and, in fact, it's using twice as much energy or what have you. It's just not well written.
Neal: Well, and that's a fundamental problem with these things if it's slurping up all the examples of sorting on the internet, I've got to think that most of the sorting code on the internet is pretty terrible. Just looking at all of them and trying to distill what most people are doing is probably not the optimum course for a lot of things that require actual intelligence and taste and evaluation, et cetera.
Mike: Well, I thought it was quite funny — there was a Reddit thread on the fact that Stack Overflow had supposedly banned AI-generated answers to Stack Overflow because they were plausible but wrong. Somebody else commented, "Well, isn't that the entirety of Stack Overflow answers that are plausible but wrong?"
[Laugher]
Mike: I thought that was pretty funny.
Neal: I think curated plausible but wrong is better than randomly plausible but wrong. [Laughs] I think that's a fundamental problem that there's been a lot of talk about are tools like ChatGPT going to take over from Google and search? One of the problems with things like ChatGPT is that you don't know the lineage of where this information is coming from, and how did they reach conclusions? By its nature, it's adding interpretation into this — of course, Google search is doing that too, just in a much less obvious, heavy-handed way.
Mike: Well, coming back to the tie into hostile tech if I was evil genius of some kind who-- if I could harness these tools to write politically influencing opinion, individually targeted at people based on their own preferences and tuned to the way that they see the world and talking to them in a language dialect and idiom that appeals to them, surely, I'd have a super powerful tool for subverting world governments and stuff like that. I'm not suggesting anyone who owns an AI company and a social media platform might be doing any of that kind of thing — but if I want to do bad stuff in the world on both social media platform and AI company, there's a lot of things you can imagine stringing together on all of those!
Ken: Yeah and that's also where — one of the bigger conversations we had around metaverse and stuff is, is the metaverse or are the metaverse and Web3 one thing, or are they two things? Or what's the interaction there because there's a lot of people inside Thoughtworks, outside, everywhere that think that they're tied, joined at the hip, whatever the idiom you want to use is, and we were like, you can't have a Metaverse without crypto payments, it's like, well, why not? We've been able to take digital payments for some time now. Again, that's one thing that happens quite a bit in my household, I think the Amazon drivers know as well. Digital payments are not difficult and so does it need to be tied together. I'm curious because Mike's much more into Web3 than I am, but at least knows more about it.
Mike: I think into, yes, into is a loaded term there!
Ken: Yes, sorry. I didn't mean to accuse you of anything untoward there. When we really looked at these and said, there's a lot about Web3 that is decentralization which might be the selling point for metaverse, but that's certainly not what they're trying to do because-
Mike: Well, I think-
Ken: Mark wants you to come to his metaverse.
Mike: Exactly. Why is Facebook building a metaverse? It's because they see it as the next major forum for human beings interacting with each other and they’d like to own that. The fact that Apple, goodness gracious, puts limits around the amount of tracking you can do to people with apps on their platform, actually, is listed by Facebook in their quarterly earnings as an impediment on their business model that Apple would put any kind of a restriction in place.
Say what you want about the Apple tax on the store and the fact that Apple is raking in money from owning a platform, what Facebook is doing is building their own thing and in their first early developer release, where you can pay content producers for their content, the meta cut on that was 50% — and, like, could they signal this any more clearly what they're-- at least be a little bit covert about the fact that you're trying to own the next major platform for humans connecting to each other and just siphon money out of it?
My point is that whenever you talk about these things, and Web3 is another one, you actually really have to look at the motivations of the person doing the talking to unveil a little bit more about the actual technology position that they're talking about. When folks say, obviously, you need cryptocurrencies for the metaverse to be successful, there's a lot of questions there about what do you mean by a metaverse? What do you mean by crypto? Personally, I think there are ways that closed, 3D embodied internet experiences could be very successful without requiring globally distributed trustless cryptocurrency transaction things. There's plenty of ways that you can do ownership of digital items without requiring NFTs, and people will say, "Oh, yeah, but then you have to rely on a single company." I'm like, "You know what? If Blizzard makes a compelling metaverse experience thing with digital item ownership within that, I can see a whole lot of people being very happy just running around in one particular metaverse, owning digital items in that one particular thing, and feeling very much like they were fulfilled by the vision of the metaverse," when other people would point at that and say, "Oh, well, where's the interoperability? Where's the decentralization? Where's the competition between the experiences?"
I think part of this also depends on where we think this is going to go: we got the internet because a bunch of very smart people gifted it to the world, effectively. All of the research, a lot of it was government sponsored. A lot of it was academic, but it was open, and that's what gave us the internet. I think the problem with things like Web3 and the metaverse is that every commercial company in the world now knows how much money there is to be made. Walled garden is the default approach and only later maybe we finally see some interoperability.
Neal: I think so much of Web3 is a solution looking for a problem. You both alluded to this — that digital currency is really about distributed trust; so is a lot of the things that they're selling — “oh it can only be implemented on Web3 and on blockchain and related technologies,” and it's not true. There are lots of ways to implement those things. That just happens to be an implementation choice.
This goes back to one of the quotes that I'm most well known for, in one of the 97 things every software architect should know is my quote that says, "Developers are drawn to complexity like moths to a flame frequently with the same result." It's just something really enticing about, "Oh, it's a globally distributed blockchain. Understanding the algorithm and understanding how it works," it's like, "Oh, we've got to use that for something," and it's a solution and searchable problem in a lot of cases. I think there are use cases for it, but most of the uses I've seen for it are actually better implemented in different ways. I think the hype train is really, really heavy with this one. I think you're both right — they've hitched it to the metaverse and some of the other hype trains that are chugging along and trying to get them to all go as a group.
Mike: Yes. As a counterpoint, though, I listened to a few finance podcasts and there's one with some fast talking New York finance folks on it. They were saying, if last year's events in the crypto space have not killed Bitcoin — which is still holding 16,000 as we record this — if they haven't killed Bitcoin, nothing's going to kill it, kind of a thing. Maybe there's some truth to that. We just don't know how these things are going to pan out and how they're going to end up.
I think part of the advice that we try to give in the Looking Glass report is around what people should be doing for all of these technologies. In the Web3 space, for example, I think our advice would be to be careful before getting involved in it, to be clear about the kinds of things you're trying to achieve, and to put a box around it and be very clear about it. There's a lot of folks who jumped on the NFT hype train, and a lot of which I think is not a very good look right now from a brand perspective to have jumped on that hype train. But that doesn't mean that it's not going anywhere. You need to do enough in the Web3 space to understand and stay close to it so that if it does take off, if Web3 payments become a big thing, you can understand that as a business and get involved in it.
Ken: I think you just hit on something there that applies to all the lenses, really, but especially these three that we've been drilling down into, is that as a business, you have to be prepared for what might happen. Just the way that user experience happens in a metaverse thing, the interactions are different in a 3D situation than they are otherwise. If you don't have any idea how to interoperate with a blockchain, if you don't have any idea how to incorporate AI into your product. There is a sea change: a fruit manufacturer out of Silicon Valley comes out with new AR glasses, and the whole market changes — you do have to be ready. There's a lot of this stuff in here that we're like, "Ah, I don't know if it's mainstream." We said is there any there that sort of thing. There's a lot of these were the answer to that's not clear. That doesn't mean that as an organization you shouldn't be preparing and building capabilities and building understanding.
Neal: One of the things that-- and Ken was just touching on this that one of the things that doesn't come out as much in the written version of this is the synergy between these things. These are treated mostly as discrete things, but a lot of what we've been talking about is the synergy because like everything in the technology world there are no clean binaries; everything's a messy spectrum. A lot of these things end up influencing each other. To Ken's point, if the metaverse really takes off in the 3D virtual world avatar sense, then Web3 is probably going to be an integral part of that in terms of payments and distributed trust. The question then becomes: exactly how do you implement those things?
I think it's important for all these things to separate the thing that you're after from the implementation of how you're getting it. A lot we get bogged down a lot in the solution versus the problem. I think some of these reflect exactly that tendency that we have in the technology world.
All right, so any last words about the process of creating these things? Anything that really surprised you during the-- as you were putting these things together? Any things that almost made it, didn't quite that is worth…?
Ken: Probably the biggest observation about the process is that this report and it's in the state if you go to thoughtworks.com and look at Looking Glass is really-- it's global. It's not industry specific. It's not market specific. It's not anything like that. And so, what in some-- a lot of companies have done, several actually, and we're here to help with this is say, okay, how do we take framework like this or a template like this and apply it to an industry or an individual company or what have you, because there are things here that we're saying, globally, this should be the tack on X. Whereas for finance — if you're a bank — that is going to change the focus of some of these lenses to use the pun. There's a lot of that where we had to go at a fairly high level on some of this, where it's really interesting sometimes. Like I said, we've done it in several cases — really drilling into it gets really interesting.
Mike: Underneath the six lenses we've got there's actually a hundred individual technology trends in there. We categorize those technology trends by how soon we think they're appearing, how far over the horizon they are or whether they're with us today. Also how urgent it is to make a response to those individual technology trends. Again, that advice is fairly broad. If you're in a specific industry, a specific company, then your interpretation of the positioning of those things might well be different. What has been super useful is for teams of people to use this as a starting point and say, "Okay, what do we think about our context? Either our company, or our division, or our department, our team do we agree with this thinking from ThoughtWorks that is industry-wide, would we modify it? Does it cause us what has Thoughtworks not included that we see that is important, and how do we want to respond? How should we respond to these technology trends that we've identified?”
The thing I like about it is it's a framework for causing you to think about things and to seek extra input from what's going on in the world and to be able to factor that into your own decision-making. This report doesn't exist in a vacuum. We don't expect people to just read the Thoughtworks report and not read the ones from any of the other companies out there in the industry. I hope that this is a good piece of input that isn't exactly the same as all the other pieces of input that you're reading; it's got the Thoughtworks perspective on it. I think regular listeners will understand roughly where we are coming from on things and how we differ with some of the other organizations in how we think about stuff. That is born out in the looking glass.
And that to me makes it a good starting point for people to consider how their world is impacted by these technology trends and also any other technology that we didn't feature that they think is important.
Neal: I think that's really key. Actually what I was going to say is exactly that. Don't take our word for it. This is a good template for going through this exercise yourself. It's very easy in the technology world to get caught up in the day-to-day of the tactical things that you have to worry about and then the strategic things start falling away because you haven't addressed this thing. So this is a way to encourage you to think about the near-term future and use this as a starting point. I think individual companies and individuals will come up with their own list of really important things to look for. I think that's a useful exercise.
That's a good time to wrap up this edition. We'll try to remember next year about this time to come back and see how we did in our predictions. Although our predictions are pretty light, we're not trying to predict a sport seems winnings or anything, these seem pretty safe. We'll see how the world has changed you. The people who created this was pretty fortunate because I think between the time you wrote the lense and it was actually published, ChatGPT blew up and became all the rage. You were exactly dead on with that one. We'll see how–
Mike: Partnering with AI certainly seemed relevant in the face of ChatGPT.
Neal: [chuckles] Exactly. Thanks so much, Mike and Ken, for your effort on that and spending some time by giving us a really nice overview, the Looking Glass, where you strongly encourage people to go read the real thing. There's a lot of information there, and some good actual advice to take away and get to see what our perspective on the near future is. Thanks, guys.
Mike: Thanks, Neal.
Ken: Thank you.
Neal: Thank you for joining us and hope you come back for the next episode of the Thoughtworks Technology Podcast. Have a great rest of your day.
[Music]
[END OF AUDIO]