Zhong Tai: innovation in enterprise platforms from China

Podcast Transcript

Neal Ford:

Hi. Welcome to the Thoughtworks Podcast. I'm one of your regular hosts, Neal Ford.

Mike Mason:

And I'm Mike Mason. I'm also one of your regular hosts.

Liu Shangqi:

Hi there. I'm Liu Shangqi. I'm a technology advisor working out of Thoughtworks China.

Xu Hao:

My name is Xu Hao, I am the head of technology of Thoughtworks China.

Neal Ford:

And today we're going to talk about a fascinating subject that came up during our most recent tab face-to-face meeting. As many of you know, we meet twice a year face-to-face to produce the radar and very often fascinating conversations and topics come up. And in fact, this time, this topic actually came up a couple of radars ago with a tough translation, and we had a full presentation on it this time because we were in China. And today we're going to talk about a subject that you can't even Google in the West, because when you try, you get something that doesn't really mean anything.

Neal Ford:

It's this concept that we first heard a couple of radars ago, something called the Middle End, and it was described to us as, "Well, you have a front end, and you have a back end, and you have the middle end," but then it was also called Zhong Tai, which is the thing that we heard about now, and it turns out there's a lot of interesting nuance around platforms. But let's let the two native folks who know about this thing, and are building it everywhere, talk about it.

Mike Mason:

I just want to make sure that we get the name right as well, because I think all the westerners who are seeing this written down are going to be pronouncing this "Zhong tai" and that's wrong, right, like it's-

Liu Shangqi:

Close enough.

Mike Mason:

Well no, let's try and get it right, so how do you say it?

Xu Hao:

So in Chinese we call it Zhong Tai.

Mike Mason:

Zhong Tai

Xu Hao:

Yeah. Probably, I can give a little bit of background. Even today, we don't even have a clear definition for what Zhong Tai is. So the idea I think goes back to four or five years ago. So Alibaba, most people know that this is the largest e-commerce website in the world. One day they announced we're going to be starting a totally new B.U. So the-

Neal Ford:

B.U. being business unit.

Xu Hao:

Yeah, business unit-

Liu Shangqi:

Business group [crosstalk 00:02:14]

Xu Hao:

So, they said we're going to start a totally new business group, so that's targeting small enterprise. They're actually talking about say ... In a digital world, the most important thing is no longer your internal system. It's more like you have to have an online channel to contact your customer to selling the stuff directly to them. So they actually say, okay, we already have two of the most successful digital channel in China online, one is Taobao, so one is Tmall, so they're targeting two different customer groups.

Neal Ford:

So Taobao and Tmall.

Xu Hao:

Yeah. Yeah, yeah, yeah. So Taobao is primarily for the-

Liu Shangqi:

C to C.

Xu Hao:

The C to C or ... Consider, like, a tiny business to customer. So-

Mike Mason:

So, like Etsy? Kind of like handcrafted goods sold online, or small-

Xu Hao:

Yeah, yeah, yeah.

Mike Mason:

Like garage business to consumers.

Xu Hao:

Yeah, to consumers. So Tmall is more like small enterprise, so basically you have a brand. For example, Nike is actually one of the customers for Tmall. You will have a much fancier shop front, you will have to pay some money to there to get to customers. So basically they say, okay, we already have those two, but if you want to build your own, you can start from scratch. Or you can build on the new platform we are going to provide, so they call them the Zhong Tai. So basically they describe the Zhong Tai as encapsulating the business model. The idea behind those is we already have a successful business model in the online e-commerce thing, so then pretty much, your 80% or 90% requirement is already covered in this scenario. You may not feel comfortable to associate your name with Alibaba or Taobao because you don't want to run your business on there, so people might say, since you're going to be tiny or you're going to be a small enterprise.

Xu Hao:

For some enterprise, they want to have their own online identity, they want to run their own channel. So those requirements for them is they're starting from scratch, throwing a lot of money in there. They say, okay, here is platform that is encapsulating the best practice for the online digital channel. How can you approaching your customer? How do customer relation management? How can you do the online customer ... online customer service? Online chatting, and those sort of thing? And how can. .. how do you integrate it, the whole thing is actually packed.

Mike Mason:

So if I was a ... What kinds of businesses would then use that, that e-commerce capability? Would that be brand new businesses, or existing businesses and they would then ... they don't like their old website anymore, they want some new thing?

Xu Hao:

It's the existing one, because the reason being Taobao and Tmall, they actually provide the standard online shopping experience, because most of the enterprise, they don't need to catch up, the online experience will be just like Taobao or just like Tmall or even better with online identity or something like that.

Xu Hao:

But to reach such a user experience, it's actually quite hard, because Taobao, it took them 10 years to actually polish the product, to have ... to today, like, the de facto centers for e-commerce or online shopping. So basically, it's the existing company, they're just doing the catching up kind of solution, because they want to catching up. Most of them, they already have an account open in Tmall or in Taobao, to ... like, reaching the customer base, but they're still catching up, so let's say we want to have our own online representative. So that is actually ... Back to four or five years ago, the concept firstly introduced to the China market, so that is actually ... So, Alibaba, so that one is actually has a really interesting impact in the China Tmall community.

Xu Hao:

Actually, two groups of people, they respond to all of it differently. So one is other internet companies. So, they say, oh, Alibaba came out a new fancy concept, called Zhong Tai, so how can I actually using those in our solution? So is that a true valuable concept, where it does something useful?

Mike Mason:

So just using the concept, not using the Alibaba platform?

Xu Hao:

Yeah, yeah, yeah.

Mike Mason:

Okay.

Xu Hao:

So then, the other side of the story is actually the traditional enterprise. So they say, oh, there's the fancy new shiny thing came out from Alibaba, so there must be something that we can learn, because they already know, they want to learn the technology, learn the best practice from the internet company to be able to compete with them, like in China market. So then there's something came out new, they're going to say, okay, can we borrow this idea, and use it inside our enterprise? That is actually the middle level, people in enterprise that think about that.

Xu Hao:

Some of these people, they have another interesting thought. They say, okay, if you consider Alibaba like a C2C company, then all of a sudden, they introduce a totally different business model into their business landscape. So by doing that, they actually wrap around some of their core business capability and sell them like a product. So some of the IT people say, "Can we do the same thing?" Because that sounds like a huge missed opportunity, because you already have your business capability, you see them every day you don't feel that is such a surprise, you have something you can do. But for people like not within your enterprise, in other companies, they might like to use your capability to do something totally different.

Mike Mason:

Okay, so we're going really fast here, and so I want to just ask some clarifying questions, right? So if we come back to the thing that Alibaba was offering, right? So they've built up an e-commerce capability over 10 years. It's not just a website that sells you things, it's the whole customer relationship thing, the [crosstalk 00:08:23] all the other stuff that you need to do. And they're letting other folks kind of build businesses on top of that. I think a lot of people are familiar with what Amazon lets you do, in terms of Amazon as a marketplace for selling items.

Xu Hao:

Yes.

Mike Mason:

How does that compare to what Alibaba is doing in this case?

Speaker 3:

So basically, it's like ... So for the Amazon one, basically, you have a shop front that is under Amazon, and they have very limited option for you to customers as they're looking for your ... But sometimes it's good enough, but it is-

Mike Mason:

It looks like Amazon, basically, right?

Xu Hao:

Yeah, yeah, yeah.

Mike Mason:

Okay.

Xu Hao:

So the problem for them is just some of the company in China, they say, they want to have full control of their online identity, what they call the digital identity, so when people go to your website, so they know it's your website. It's not like in Amazon, and-

Mike Mason:

Right, so we might call that white labeling.

Xu Hao:

Yeah, yeah, yeah [crosstalk 00:09:27]

Mike Mason:

So they want to white label Amazon, which Amazon doesn't let you do, as far as we know-

Speaker 3:

No, they don't ... So basically, that is one thing. So another one is for the different type of product. They may actually require different capabilities. Some of the enterprises say, okay, I wanted to do a pre-sale for something, they want to provide much more customization, like, for their customers. So basically, it's not exactly identical for each one of them, because they ... We still need to know you have your online thing, and there is another really big platform there, is called WeChat, then probably so you're going to think about how can you integrate those content and your ... your inventory system to the multi-touching point?

Xu Hao:

The whole thing is not only, say, I want to, like, support you to give you a e-commerce website, but it's a several capability, you can do to managing the whole online experience. For example, if you have a Taobao account or you have a Timel account, you may have WeChat is another touching point, so all those things are actually managed by an application. You build on top of the Ali Zhong Tai.

Liu Shangqi:

Another question is for the China market, especially for the IT enterprise, is not as mature as the western side, so a lot of small and medium enterprises trying to catch up with the IT part and they may lack of the ERP pieces of lack of SAP and the Alibaba's Zhong Tai just provides their whole capability for how to manage your stock, the human part, how to manage your shopping category and how to integrate it with their payment system, and you've got this whole system and the way's out to building your own IT solution. I think that's why a lot of small, medium enterprise chose the Zhong Tai.

Xu Hao:

So that's the interesting standpoint because that is actually the first time Zhong Tai introduced it to the world, so it's wrapped around, say, it's not a front-end system just like a e-commerce website. It is not a settled back-end server you buy and install on your own machine, because it's a wrap-around, more like an ecosystem. They have a subtle capability rather than one or two, so those actually encapsulate the whole business model in there, so that's why they give the term Zhong Tai, or, just we try to translate it as "middle end," it is itself like a end, but it's actually not in that case.

Xu Hao:

The interesting part is actually the other internet company that takes this concept and try to apply the same thinking in their context. Then they actually help us out to understand better what is Zhong Tai actually is. So one of the interesting case is actually DiDi, so that is the Uber equivalent, or they actually brought Uber in China four years ago, so actually they-

Mike Mason:

Right, so DiDi is the Uber competitor who actually won in China, because I guess they just bought Uber in China. Uber said it's not worth trying to compete, we'll just let DiDi-

Xu Hao:

The interesting answer is after DiDi acquired Uber, they actually have two settled systems. They're pretty much done the same thing, but Uber is more focused on self-driver, you register online and then you pay using their software to help your customer. DiDi itself, it's actually starting from the taxi driver business. So actually they installed app on the taxi driver's side, so they're using those things and then they're gradually moving to different market to allow the personal driver in some other market. So they actually ... When they acquired Uber, so there's one more system pretty much that does the same thing, so they have an argument. What should I do? Should I converge them into one, or ... but it's a totally different product team, because they don't have any intention to merger the business unit together, because the Uber, like, a standard brand. They can still there, in the China market, so when foreigner come to China, they can still using Uber to call a taxi.

Xu Hao:

They have their separate operating team, so at least at the very beginning, so they have their separate operating team. They have their separate product owner for, like, whatever product they want to build. On the other hand, they have their own product to do pretty much the same thing, so about the same time, when Zhong Tai, this concept, came into the market, they said, okay, let's try to using Zhong Tai, this kind of concept thinking, can we using the same concept to build a Zhong Tai, to actually enable so we have commodible product? So build on those?

Xu Hao:

They actually figured out there's only a handful capabilities that you need to build an Uber-like services. So one is the resource management, so basically let's say if you're a taxi, or if you're a like, whatever, so there is a resource management, so there is location-based user requests. When a user want to call a car, so you just call it. And then there's matching engine, so just match those two. There's a payment, there's membership management. So basically, there are three or four core capabilities. After you have the sort of four core capabilities, you can build a product pretty much the same. So basically the idea is actually pivot your core capability into different market [segmentation 00:15:20].

Xu Hao:

They actually figured, oh, that's actually pretty interesting because we can using one Zhong Tai to actually enable multiple products, so all those products is actually going to helping us to targeting different market segmentation. One of the interesting story after they merged into this Zhong Tai strategy, they did start a new business, which is called the Mini Shuttle Bus. So for example, in China, we have huge residential area, so in the residential area, some people, they may go to the same place. For example, Software Park, because they're working there. They all sending requests, saying, we want to have a shuttle bus just run from our residential area into the software park. So if you summon this request to the government, they're going to take years for them to actually set up a bus, because they ... they have to, like, evaluate how much cost are they going to put in.

Xu Hao:

But for Didi, they provide this service, they say, okay, if you have enough people, if you pay monthly, if you pay $100, then if you can find 30 or 40 people can pay $100 per month, then we'll send a shuttle bus for you. Some people have to hire a driver to drive a really huge bus to just run that route, so that is actually a really special requirement, and they only put those into certain areas, like Beijing, Xian, because they have pretty much from residential area to software park, because they have huge people that work in the software park.

Mike Mason:

But do you still do that from the DiDi app?

Xu Hao:

Yes. You need to have a separate ... they have a separate form, say, I want to send me those, if enough people say, "I want to have a mini shuttle bus," or something like that, and then you will see on your app there is mini bus going to come to your location. Five minutes later, it's going to come, or something like that.

Mike Mason:

How dynamic is that? Is that each day, DiDi is choosing whether to [crosstalk 00:17:14] a bus?

Xu Hao:

It's actually a yearly base, because if they decide to have the route, it's going to run for a year.

Mike Mason:

Okay.

Xu Hao:

So they have to make sure they have enough people to actually run those. They actually say they want to redo them and send them monthly or something, but today, they still, like, yearly base. So this idea, and then they pivot all their existing capability to build another product, particularly for that customer segmentation, so then people realize, oh, Zhong Tai is like encapsulating the business model, which can enable you to build multiple product for different market implementations. That's actually one of the popular concept, or definition, for Zhong Tai, is actually from the learning, from other internet company.

Liu Shangqi:

Yeah, I think the key point here is to encapsulate the business model into your platform, into your Zhong Tai, into your service, because, you know, for techies like us, always think about how to encapsulate your technology platform to build up your platform for your [inaudible 00:18:23] delivery, for your operation, for your cloud computing, but another side of the word for the business man is you think about how to replicate your business model, how to ... how to do that, replicate that. I think the Zhong Tai idea is to merge these two ideas together.

Speaker 4:

If you see a platform called Zhong Tai, but it's just focused on pure technology part, it is not the Zhong Tai we're talking about, another concept. But if you see a company, they may trying to leverage their business capabilities to provide maybe API or service and to support different channels, different products, different business units that may ... They may not know that Zhong Tai comes out, or they may use the right way of Zhong Tai.

Neal Ford:

That to me sounds like the difference in... So Thoughtworks talks a lot about platforms and platform thinking as a way of encapsulating technological capabilities within the enterprise, sort of like a higher level of componentization, but that's always about technology assets and technology APIs, and it sounds like Zhong Tai has a lot more of the business model baked into it. And it also sounds a lot more sort of turnkey, where you get a lot more from the platform than just technology assets. You get business workflows and sort of a better whole package.

Xu Hao:

Yeah, that is actually interesting, because ... the idea from Zhong Tai, I think that's actually really close with the cloud technology, because I think... So I don't know the real story behind those, but if we reason back why people want to do that, it's kind of really interesting, because Ali itself is a cloud provider, so they provide cloud-based structure.

Speaker 3:

In the cloud, the most easy and cheap thing is actually to replicate. So if you have something in the cloud, the most easy and the most cheap thing is actually to replicate the whole thing.

Mike Mason:

Sort of like cut and paste?

Xu Hao:

Yeah, yeah. [crosstalk 00:20:34] It's pretty much is like that. Then the Zhong Tai came out, is actually the whole platform thing is actually not a platform, it's more like a SaaS of a platform, because from those you copy-paste the whole platform and put them into different location. So even though it's happening at the service level, you can have the software service, but concept speaking, so it's actually, say, you can copy-paste your platform and [inaudible 00:21:05] into another team or to supplement another product, like to build.

Xu Hao:

So then it may inspire some people say, we always try to find a new business model, but if you already have an existing one, can we run a business? Can we earn much more profit than just replicate an existing business model? Different people to operating on the same product or a pretty much similar product, but two different market segmentation, is that the way? So we're going to go to the new market? Or can we using both, because that is technically the most cheap thing to do? Can we using those, kind of thinking, to shaping the way with these other product. Because the most cheapest ... so the most cheap, the most fast way to do that, is you already have a business model, you replicate that, and you find a customer segmentation or market segmentation for that.

Xu Hao:

You have a bunch of people building your product, and then you go there, do whatever you want to do. So I think it's not ... speaking from a business world of thinking, because it ... So the thing for me is purely because today in the cloud world, it's really cheap and easy to do replica, so if ten years ago, we wanted to building the same idea, it's going to be a nightmare, because how can you actually replicate a really complicated platform and handle those two different teams to allow them to build their product? That would be super different, so this one, for me, is a really interesting case study for how this technology choice actually impact the people thinking about business.

Neal Ford:

So you mean it's hard to copy and paste servers?

Xu Hao:

It may not be, because you can run it as SaaS, right? Because SaaS basically is copy and paste, but in a much more manageable way, because we don't do the monitoring, so basically if China should just copy-paste, or you do whatever funky thing you're going to do behind, but the essential idea is going to say, I want to have a service, I want to have a thing, and then I will make it available to multiple people. That is, by definition, copy and paste, right?

Neal Ford:

Mm-hmm (affirmative). So this, it strikes me ... So back in the late '90s and early 2000s, IBM embarked on this terrible, terrible experiment that they called the San Francisco Project, and the idea was, build one last general ledger system, one last inventory system, and it was this big, giant, big design up front experiment that failed utterly because businesses are so complex. But this sounds almost like a reasonable implementation of what they were shooting for, and it actually uses something that we've talked about, which is harvesting frameworks rather than trying to design them up front, because it sounds like what people have done here is harvested really useful e-commerce bits and all sorts of IT bits at the business level, and then packaged them for use and are calling that Zhong Tai. Is that an accurate assessment?

Xu Hao:

Yeah, pretty much. The idea is actually they're building out microservices and they actually identify the capability rather than a whole package of software. So basically, every Zhong Tai I've been encountered is actually a settled microservice so that is represent different capability. There is a kind of, like, if you do in a certain way, it's much more easier, but there is certainly no ... enforcement saying you have to do in one way or another, because it is just a bunch of API call. You can call your own API from your own Zhong Tai, but you can call the third-party API for payment, for example. So those are actually pretty easy. So it's actually conceptually is an encapsulation for the business model, but from the implementation perspective, it's a still a set of loosely-coupled microservices.

Mike Mason:

And you mentioned the Zhong Tai you've actually encountered.

Xu Hao:

Yeah.

Mike Mason:

That sounds like there's more examples of this than just Alibaba, right? Because that's the other question. When you come in something like this, are people really doing it, or is it just a good idea?

Xu Hao:

So it's actually two type of response. The first type of response is actually for the other internet companies, like DiDi or Baidu or Tencent. They actually say, okay, it's actually a good idea because they're more familiar with the technical side, so they're more familiar with the call side. They actually have a lot of common things inside already, so they say okay, just try to find a different way to encapsulate the whole thing and make a Zhong Tai for different purpose. So that is actually ... They were actually quite positive about this one, because otherwise if only Alibaba or Tencent say it is not a good idea, they're going to come out with their own version of this thing. And people say yeah, it's a reasonable encapsulation. The internet companies, some of them, some of the really big ones, they actually rebuild their core systems more in the Zhong Tai way, but they may have multiple Zhong Tai rather than a single one, given their size and different area, so they may have slightly more.

Xu Hao:

The interesting idea from that is actually ... so the interesting response for that is actually from the other enterprise. So it's not the internet company, but it's from the more traditional enterprise. They were treating Zhong Tai somehow, not only like a strategy to encapsulate your business capability but it's more like a lever for the organization transformation. Because they were being encountered a really pain in ass, it's actually how can we build an autonomy product team to have a long life product? Like, rather than a project? Because your business team say, I won't do anything, then you start a project, so after project done-

Mike Mason:

The team goes away [crosstalk 00:27:18] long lived thinking-

Xu Hao:

Yeah, so they really struggled, say, how can we actually build a product? They learn from the internet company, if you build product, people will have more ... They feel more responsible. They actually have more idea, there is more innovation. The quality is going to be better. There's all sort of benefit from doing that. But for enterprise, that's real difficult to do a large-scale transformation to build each single product. So they actually feel if they build Zhong Tai, it can be a good leverage or in the middle step for them to transform from one to another, because by definition, Zhong Tai is encapsulated and going to be replicated ... like, being smaller. So usually, it will be done in a self-servicing SaaS kind of platform than there.

Xu Hao:

The idea is saying, if I build a Zhong Tai, then you have to have a long-lived entity. Say, I want to have a lease on your Zhong Tai to have my own platform, there is going to be a contract between your Zhong Tai team and your product team. From there, the entity who actually leasing the capability from Zhong Tai is now going to be the product team, because your product may now exist next year, so from internal business collaboration perspective, so they have to find a longer-lived business entity, say, okay, how about a year or two using our technology to building our own thing? So even though we don't charge you, but for your own KPI or whatever, we want to ... More formal contract way.

Xu Hao:

Then they actually can using those like a lever to actually build a smaller product team that is more longer-lived, and then say, okay, almost all of the thing in there is actually building the capability, like you need to build a front end. You need to do customization, and you need to do operation. Then you will have a self-operating team building product there.

Xu Hao:

Then they feel, even if we can't find a business capability that is actually replicable, like a [inaudible 00:29:19], by using those kind of thinking, we can actually make our own organization structure be much more better. So that is the actually more practical thinking from the petitioner. So they say, okay, maybe we are not going to lean on a company to find a business model encapsulating that, and replicate 100 times to 100 different market segmentation, but this is actually a way, since it's so hard in the tech community, so the business man actually buying those ideas. It sounds like a really good idea, then they're using those to help their organization to become a better organization.

Liu Shangqi:

And it's interesting that if you see the internal organization of Alibaba, they are a lot of ... a lot of the business architect, their background is technical people. The major lead for each Zhong Tai shared service majorly there is a business architect and most of them come from the technical side, so they were ... The bad part is they were technical architect before and they are doing this self-service regarding their service, not just a project but also a product, and they're learning the domain knowledge a lot, a lot. And they learn is there's no one in the world that knows us, why should management know the inventory management better than they become the domain expert. But before that, they were pure technical expert.

Liu Shangqi:

By doing this, it is one way to enhance their responsibility for their business, for their service, but also, I think, make the more focus on their specific domain to make them the domain expert.

Mike Mason:

So is there anything about this that, I know when we talked about it a couple of weeks ago, it seemed like maybe this is happening here because of the scale of China? Is that a factor to talk about?

Xu Hao:

That factor is actually ... So when we talk to our customers, say, okay, you need to find a business model and replicate that. So the people say, okay, there is only a handful of market for my domain. For example, banking or financing, they say, okay, it's really difficult to think about, I want to using those banking capabilities if I want to do something else. So it's designed for banking. So you're going to ... like, expect. So the idea for that is interesting. When we talk about market segmentation, it's not necessarily mean new market, it also can mean the existing market, but you can tier them down by different criteria, like, for example, you can build a product for the Tier One cities like Beijing, Shanghai, Shenzhen, those really big Tier One cities. You can build a lot of product with exactly the same capability for the tier four city. They have different needs. They have different product. They require for different product.

Xu Hao:

So the original way to do so is actually putting our thing in one piece of software that's super complicated. So even though those are using the remaining platform behind, but the software itself is just overly complicated, because once we know the [inaudible 00:32:34] from one problem they have 4,000 different product, like you can buy for the data whatever, they combine everything and it's, like, 4,000 there. In a huge province, like in Guangzhou, there's tier one city, tier two, tier three, tier four. So they actually require different type of service. So the service makes sense for the tier one city, it doesn't make sense at all for the tier four city. And likewise, because a tier four city, they may be more cost-sensitive, so the product probably there can't make any sense for a tier one city. By using this Zhong Tai thinking, you actually see, okay, they are both leveraging a similar business model. Then you can actually divide your product team form one huge product team into four.

Xu Hao:

So then in that scenario, because even though for the tier four city, there's still millions of people there, so it's worse to build a product team that is actually responsible for a tier four city, and then you can have a better service. If anything changes on the market, you can be response much more quicker because otherwise if you're putting all this marketing change and putting into a really big backlog, there's no way for you to actually response quickly enough to the whole thing. But then by using this Zhong Tai thinking, so you can actually divide an existing product team into a more responsible, more like a product design, because usually you won't mess up with a really large-scale application into a different type of thing. That is actually kind of like changing the way people think about how they design their product.

Xu Hao:

Is it nicer to have a product to serve all the customers, or it's going to be much more easier if we can just simply divide those into much more smaller, or even different type of products.

Neal Ford:

One last critical clarification question. Is Zhong Tai a noun or a verb? Do you ever say, "Let's Zhong Tai this," as a way of pulling apart, or is it always used as a noun, saying, "We need one of these in our organization?"

Xu Hao:

For me it's more like Google. It started as a noun and then it turned to a verb, so it's starting [inaudible 00:34:51] in China community, it's kind of turning into a verb now, but it's still not there.

Neal Ford:

That's perfect. It's a fascinating concept, so thanks, guys, for describing this to us. It's a fascinating way of thinking about re-use and it's super fascinating to me that, as homogenous as the technology world is, that there's this whole branch of technology thinking going on in China that's largely opaque to the western world. Thanks very much, guys, for giving us a fascinating insight into it.

Xu Hao:

All right.

Speaker 4:

Thanks, bye.

Mike Mason:

Thanks, guys.

Rebecca Parsons:

Coming up on our next Thoughtworks podcast, we will be talking to Biharck Muniz Araújo, who wrote a book on REST and Typescript and Neal Ford and I will be talking to him about his writing experience as well as his choices and so we hope you'll join us for the next podcast.