Has Software Eaten the World? A Discussion on the Evolution and Future Direction of Software

Has Software Eaten the World? A Discussion on the Evolution and Future Direction of Software

Jeff: [00:00:00] Hey, Brian, how you doing today? Doing great. Jeff, how are you? I probably ate a little too much candy this past weekend, but other than that, I'm feeling.

Brian: We have, we've all had that problem. I think we need to have a service that would take Halloween candy back.

Jeff: That is something that there needs to be a start-up that comes to your house about about 8:05 PM on Halloween night and take the candy

cause unfortunately that bag is still one of the bags is still in the house. That's a really good idea, Brian. I like candy as a service.

Speaking of eating, 10 years ago, Brian mark Andreessen wrote that famous wall street journal article.

Why softwares is eating the world. I'm curious now, 10 years later, what do you think about that? We are where are we with what mark was saying?

Brian: I think it's a very interesting question. Yeah. A lot has happened since then. And I think, there's a really interesting quote from, what Fred Wilson, the very popular venture capitalist from Union Square Ventures once said, I think it's been said by many people.

I think Bill Gates said it, previously as well, [00:01:00] which is like people always overestimate what can happen in a given year, but underestimate what can happen. In 10 years,? So it's almost 10 years since that quote came out and a lot and lot has happened in the software industry. I think we've come a far away , especially with low code and AI, IOT, all these technologies being abstracted by software and also the ability of software propagating a lot faster. In traditional industry, because, it's obvious Silicon valley has been in this for the last 50 years in software. But now we are seeing the propagation across a lot of the traditional industry players. All the way from beauty to insurance, to automotive, to, convenience industries that are, that were very manual before using novel users of software, not just simply digitizing the customer touch points, but also using AI IOT and machine learning in a much [00:02:00] more effective ways to understand consumers better, to understand interactions better to.

Manage their inventory, do a lot of different things that are super interesting. The touch points, all using machine learning. So I think software has come far away.

But I also want to take this moment and talk about two interesting paradigms. In terms of evolutions, the first paradigm is basically the software evolution stack, which I call it.

A lot of different leaders have spoken up about it in different ways. What happens, the, what this tat means is it explains, how software has evolved over the last, every decade, how it has evolved . In the nineties, we've had packaged software. So there are so many companies, providing, like Lotus notes and all these companies providing software that was packaged software.

You go into a software shop and buy packaged software, and even digital versions of packaged software, all the early players, Oracle, and many others provided packaged software. But then [00:03:00] time passed. T he installed package software on the layer two or the next evolution became SAS.

In the early, in the late nineties, it was called application as a service or application software rent for rent. That's what it was called. But over time it became software as a service where customers did not own the software, but they took advantage of the software that was hosted and running.

In the servers of these software providers. So that's the, so that SAS paradigm is still there. A lot of our software that we use every day. If you like, even in our own company at iterate, we have over, I don't know, 30, 35 SAS windows we use on a regular basis all the way from accounting to, software development everywhere, throughout the different parts of the organization.

In a traditional organization probably use users in average, over a hundred vendors, ? All SAS vendors, providing different parts of software across many different bias of every [00:04:00]department. But then the third evolution is component software. So now what happens is it got more granular as any industry.

Things get more granular, ? Just so the component software is essentially like a supply chain mechanism. So today, if you want, if you're looking at a car, the, there is a manufacturer, there is a tire manufacturer who makes the tires . There is another manufacturer who actually does the stereo system and other one that does the doors and the interior.

Similarly for software there are very specicialized software manufacturers like Twilio is a classic example. They provide communication software and components software? In the AI space, there are a lot of players who provide AI recognition, capabilities, and various software components. So that is the second the third evolution of software where everything is componentized.

There is somewhat of a virtual supply chain mechanism, and then as engineers and then leaders within companies. Putting out these digital initiatives, they assembled software across these paradigms. That's the paradigm. So today we have that is the [00:05:00] paradigm that's in action today between that and that, and the older SAS.

Those are the two kinds of paradigms that you see in action today. You don't see very many packaged software anymore. But then the paradigm that are going to talk about next, the fourth evolution, which is going to be the self organizing self writing software generating software,

not just human general humans generating software, but software generating software. I think with the, with with the growth of an and stuff happened, that's happening inside machine learning. We are not far away from. We are already seeing in a lot of areas where code is being generated. This whole idea of code being generated has been there for years, but they use traditional methods to generate.

But that's changing now with machine learning, where you can generate code for different interfaces. All you have to tell that the machine learning system is I'm interested in this type of an app that would interact with the customer this way. And these are the features I'd like to have, and then click on a button and it'll generate code, or you could, or a designer could [00:06:00] design a user interface. Put all the buttons, click all the actions, tag all the things that he wants to do and say generate code now. So this generation of code I think is interesting and also machine learning ability to doing machine driven code generation, machine driven DevOps, are

going to get very interesting. But I think that thing is probably the next paradigm. That again, So that's one dimension of the software stack or the evolution paradigm. There's another dimension, I think, which is very interesting. The software can eat the world only when software has the tentacles.

Because software is like the brain, ? The human has the brain, but without your arms and your body, you can really put a lot of your thoughts into action and it's very hard, ? So in the old world, we had, keyboards and printers that are input tentacles and output tentacles for software, in a modern world. We have visual, we have. Touch screens. We have, multitouch we have all kinds of interesting interactions with this type of, virtual reality AR interactions through, bone conduction through [00:07:00] glasses. So these are new tentacles, ?

From a human computer interactions, we are getting newer and newer tentacles with VR sets brain devices, brainwave, sensing devices, all that type of stuff. The different traditional industries are also getting different tentacles. Let me give you a classic example in the beauty industry?

In the, in, in the olden days, if you were to do a makeup or a lipstick, you have to actually go get that lipstick manufactured. Then in a manufacturing facility. Now that manufacturing can be done in a small lab, as micro manufacturing. But there are so many startups and companies working on where you can set a color of a lipstick or a set the color of a makeup or a foundation.

And it can be built for you out of the gate, because this is a tenacle that takes the color combination in the software and turns the software into actually a piece of. So these are the type of tentacles in every industry. You are finding tentacles like this that is specific to the industry, ?

So as software is getting powerful powered by tentacles hardware, tentacles, which are [00:08:00]becoming cheaper and easy to reproduce. So that's another dimension of why software related work. First is the first and is that demention has the evolution paradigm. To conquer the world and convert everything that's digital and everything that can be done.

We are a human process. We are software, secondary software is getting arms and legs are powered by these tentacles, which is powering software to get even stronger and to be, and to do more things that that we have not imagined would have not imagined.

Jeff: So that, that's really an interesting way of thinking about that.

 I assume you're making the case that these tentacles are really almost IOT's various little devices that are.

Brian: Yeah, absolutely. They are IOT is, are industrial devices. They are, even non-internet connected devices too. They can even be like non-connected devices, but they're connected to the software on that age.

On that case. Yeah, absolutely. Yeah.

Jeff: So we're well into, a service where. I'm not sure. We're really, you'd have to tell me if you can really even come up with a couple [00:09:00] of good examples of self writing software. So where are we in those, in this in this evolution now,

Brian: see, I think where we are, Jeff is we're in the majority of SAS, ? SAS has been . And then venture capitalist investors, startups, they, the paradigms around stats, SAS, ? How to build a SAS business, how to attract consumers, how to drive traffic to it, say how to optimize it, how to deliver the service has been very well understood.

So we are at the very top peak and then what's going to happen so we are in between the, the two and the three, which is component and software? So there are a lot of software today.

That's been Britain is being built by frameworks and. By using components. But what happens is only a very few companies that have been very successful in becoming component related software companies. But but that will change, but there is but one of the things to realize though, is if you look open or the open source movement in their entire world, everything is component software, ?

The open source is all about, taking a framework or a component from here, taking another thing from another place [00:10:00] and assembling. So I think we are in between the paradigms SAS and the component software

Jeff: is component software. Is that low code?

Brian: I know, I think component software is more broader than Low code software just simply means, once one part of the software, not all the software is provided by one company, ? The software has many different functions and for some of its functions, it's using a software from company a and for another function, it's using software from company B and they're either licensing these individual pieces of software.

And as a software developer or as a company, traditional organization, not a Silicon valley company, you are putting together these disparate pieces of software and the place where you are building the dock, where you are building. Putting assembling or putting together this piece of software.

Oh, interesting. So

Jeff: we used to call that just a best of breed strategy with a whole lot of SIS doing all the integration work which you're doing is [00:11:00] still, we're still talking about a best of breed strategy but massive savings around the integration.

Brian: Absolutely massive savings around the integration costs and also every different player in all the provider for the component software, they are experts on those things.

So they'd given you the best of breeds and they're all the best of breeds are abstracted in your local platform. So you can just like moving Lego, you can move these things and try to connect them and start building software and taking them to market a lot faster. So that's why the low code, no code world is very interesting.

But what's also super interesting. Jeff is, as you're having this no code, low code capability the apps or the solutions that are built in no code, low code is essentially. Solutions that are built by connecting the disparate level disparate blocks or software components. So because it's easier for for any developer, even without a lot of [00:12:00] training, being able to manipulate this type of software, it also becomes very easy for machine learning systems to understand.

Because you have abstracted a lot of the complexities within these blocks. So these paradigms can be trained. So you build an e-commerce platform on top of using a low code engine. You can turn, train the machine, how to regenerate code for any e-commerce platforms in the future. By just inspecting the connected paradigm of the existing.

E-commerce platform. That's all built on top of local. That's why this evolution is really interesting because it's nicely teeing everything up for the machine to learn it gracefully and regenerate.

Jeff: So is that, what was that? What.

Brian: So no code. No, no code is very simple. No code is the difference between no code and low code is no code means when you are assembling these blocks, you are not writing any, you're not writing any traditional programming language in the sense.

You're just assembling a bunch of blocks and [00:13:00] putting that into there and building and deploying. You're not writing any programming language, ? You don't need to know any programming. So that's no code, like for example, the web designer, the web builder, you use to build This Week in Innovation website.

That is a no

Jeff: code or low code. So I saw it's actually no code since I've literally programmed nothing, then whatsoever,

Brian: you don't have to write any code. So the problem with that is it's great, but it also has its limitations because when you want to do things that are more.

And then customize it. You have to write code, but those platforms are designed for, they are single purpose platforms, there. They're designed to do certain types of websites or certain types of screens. But when, but the platform like Interplay, for instance, the platform that we have, it is. It's a low code platform.

So you could do quite a lot of things by just drag and drop with with low code, with no code. But if you want to customize and do things deeper, because most enterprise applications and enterprise things, activities require significant [00:14:00] customization. So you can still write code and customize it. And those code, the code you write, get attracted into these blocks and then reuse over and over again. So then they become a part of the no code in the future.

Jeff: How big is this going to be?

Brian: I think this can become very big. Recently I was reading an article. It's completely on a different industry by of course, professor Scott Galloway, he called the content creation paradigm, ?

The Netflix has content creation. Infinite opportunity, ? Infinite in terms of scale, ? Because people sit at home and watch content on LA all day long, and the size of the pie is becoming very large, . Same parallels is there for low code and no code .

Even though the published market sizes, I think it's around 200 and 900. By 20, 24, 20 25. That's the standard, analyst published paradigms. But I feel these markets can become very large, ? Because it's not just about [00:15:00] no code local, ?

Just the way people have this insatiable appetite to consume content, people also have insatiable appetite to consume. Look at how much time we spend in Facebook depending on what that age is, or Instagram or whatever, any of those things, like people have, consumers are sucking all these things like a sponge, just like just sucking the whole thing up really. So the paradigm is getting bigger. The pie is getting bigger. I can give you like a very high level, the analyst prediction of, some 209 billion by 20 24, 25, Microsoft made a prediction within the next five years, there's going to be like I'm paraphrasing, of course, there's going to be 500 million apps are built and out of that 450 million are going to be low code.

No code. And then. It's going to be much, much bigger, ? Take AI for instance. If you look at AI, ? You and I have talked about this Jeff in the past, ? If you look at the AI prediction, I think initial prediction of, I think it was 180 [00:16:00] million, 80 billion, then it went to 300 billion.

And I think there was a recent IDC or one of the predictions are almost at 500. That's because it's bigger. It's way bigger than we initially thought it costs. Because as these growth markets and these new opportunities are actually far bigger. And the other thing that's also interesting about these forces is this, we always talk about the Five Forces of Innovation, AI, IOT, blockchain, data, and startups. These forces are not just by themselves. They are all intertwined. And then they are all intertwined with low code because with low code you can build on them, ? So like they all feed into each other and that feeding and that frenzy creates a bigger.

 Opportunity, ? If you look at the digital transformation market, that alone is around $1.8 trillion, ? So inside that AI coming with all the new latest number upgrades is around 500 billion, ? And then you have low code capabilities around is around 200 billion, and then you have.

IOT in [00:17:00] the in the, somewhere in the, close to, close to a trillion. But the thing with IOT, you have to be careful is hardware respected. A lot of that cost number is hardware, ? Because of the sensors, and data itself is another, 116 to 200 billion billion.

That's just because of the nature of data and blockchain is around 39, 40 billion, but that's just blockchain software not Bitcoin because that's in the trillion range. So you're looking at a very large markets, all coinciding together, and then even that large market in fathomable, large markets.

I see you're talking about that itself, I think is still. I think it's very similar to the content argument. It is it's infinite. It's. And because consumers are sucking it up, like a sponge, traditional organizations are doing more and more of it. And it's so it's propagating within traditional organizations.

And then of course, as a lot of the high-tech and the Silicon valley vendors, they are doing it. And then it's also going into, everyday objects and everyday things that we interact with. So it's [00:18:00]growing in every dimension.

Jeff: We in the analyst world do not have a way of calculating infinite market size.

So I'm not sure what we're going to do there with that number, but I guess it's safe to say it's a pretty big space or pretty big potential that being said, Brian. So what do retailers do about this? When we start talking about giant market sizes, then that means some or a lot of people are doing something.

What's your retailers be thinking about today and I don't know, whatever, wherever you see adoption really begin to take off for this.

Brian: That's a great question. Yes. I think retailers are already beginning to see this, I think, the lot of retail partner, retail players that we are speaking to, or be spoken to, or we have, deployed solutions.

They are thinking about this. They are seeing its benefits in front of their eyes. We've got the challenge in retail is a very complex industry. You mean, as because there's a lot of moving pieces, you have the brick and mortar, you have the digital, there is, and it's also a thin margin industry too.

[00:19:00] So you have to be very cost conscious in terms of what you're doing. And also move fast because digital is moving fast with, especially with the pandemic, the. The 5% digital is changed now. Now, if you look at companies that 15, 20% of the business it's changing faster, ? It doesn't mean stores are going away.

They are there, but they're also creating, they're all getting intertwined, it becomes a connected paradigm. So what's really interesting in retail is I think the retailers are seeing cases. Using low code and how quickly and how easy it is to get to market. And they're seeing its benefits.

And I think what's also really interesting is retail is also about an industry. Like you said, one is best of breed. The other is reusability. And best practice, ? Because retail is good at those things. What I mean by that is they always try to figure out how to buy best of breed solutions.

In the old world, they used to buy big one-stop vendors, and in the last five, six years, they've transformed the buying and looking at best of breed. The other type is [00:20:00] best practices because as you go look at NRF and all the other conferences and people attend to. The retail leaders want to learn from each other and figure out how to use the best of breed best practices, and also to figure out how to do reusability.

How do we really I think low code is really interesting because low code basically honors all three of . It's best of grades because it's a dock where you can bring any, all the best of breeds component and build your ship. You can dry dock it lot faster. So that's really interesting about best of breed.

The other thing that's really interesting aboutlow code allows you to use best practices because every component is a specialized component. So they do already using best practices and any are putting together in a way where you can reuse it and it completely reusable, and anybody can change it very fast and go to market.

So it's very inline with a lot of retail. That that's been out there for years like that learning and that knowledge that's been around in retail for years. But let me tell you a little bit more, let me like, drill down into [00:21:00] a little bit more like one or two levels lower and tell your use cases, ?

If you look at, if you look at. Like in the auto industry in retail, retailers was selling auto parts, auto products, auto related services, ? You can use machine learning to identify vehicles that come into your service centers, really understand using computer vision understand, damage, understand potential problems and proactively.

And have that ability to proactively let the customers know our future repairs that need to done to be done, or even to do very intelligent estimates. You can look at a given problem and use machine learning to do very intelligent estimates. And then when you go to this store and then you add a mechanic, looks at it, there is not a lot of surprise, saying, okay, you did an online estimate and told me it's 500 bucks to fix it.

But when you go to the store, $5,000 bill. That's a massive gap. How do you make that gap? Smaller, using much more deeper machine learning. So those are really interesting cases. And I, as I mentioned to you previously in the beauty industry, ? In a, you want to, you [00:22:00] want to be able to produce makeup that matches all type of skin.

What if you have the ability to really understand somebody's skin tone and actually produce makeup really rapidly, and that's using maybe a device, but the remaining is all done through software and machine learning. How do you understand. Like based on prior skin tones how do you understand that this given person, this would be the best message, the best color for skin tone, ?

And that type of thing. These are all very industry specific, ? Same thing applies in the. In the convenience industry, you have a convenience operator and you are looking at, vehicles coming in and you're able to sell them in at the forecourt. And then do recommendations of products.

They would like by you looking at people in the forecourt, you can understand what they want based on your database, based on prior purchases. You could even use their vehicle as an entry point to, to actually buy things. Because once you recognize the license plate, you can actually use that as.

Or throughout the entire source store transaction. So there is much more intelligent touch points, ? It's up to imagination, in terms of all what you can do, I was [00:23:00] talking to somebody else and they were mentioning in the an example in the past where they've done in the postal industry, where they are able to intelligently route mail. And then I'm by just visually inspecting through it throwing machine learning and then of course, in doing traditional retail, we are seeing the brick and mortar. We are seeing. The Amazon go paradigm. There are so many companies inventing a much more effectively using around all these paradigms, ?

So these are all like examples of machine learning and examples of modern technologies. But what's really interesting is as you want to put applications out for consumers to use it, the integrations that retailers can do can all be through low code because you can just call these API directly. Connect them, you can deploy an app very fast, very quickly.

I think that's where our low code becomes very interesting, go to market really fast, test things out. The other problem, I think it's also really interesting about retail is retail has a predominantly from a technical skillset, has a lot of engineers who are traditionally web engineers because they are to build a [00:24:00] retail website, the commerce site, they don't have very many AI engineers. So if you look in the world today I think there is 25 million engineers who are trained on some farmer, some or the other form of web development who can build web software. But there is only about 30 or 40,000 engineers who know machine learning.

I think there is about a hundred, 300,000 people who are researching with academics and everything, but purely from an engineer perspective, the new numbers around 30, 40,000 machine learning engine. So what happens in an industry like retail, if you will, low code where people can drag and. And all these Legos with it's an 80 20 equation, ?

With 20% of effort, web engineers can drag and drop and build very simple systems using machine learning without a lot of deep knowledge in these systems because you abstracted all that intelligence and that skill. Wow. That's quite going to be

Jeff: powerful in retail. Interesting. So I assume the enterprise software players fairly.[00:25:00]

It can fairly easily migrate into this new paradigm. They just, they create functionality. How it's deployed on a low code platform is probably straightforward. Relatively speaking, it's fairly straightforward, but what about the cloud players? How do they fit into this? How does AWS and Google.

Brian: That's a great question. See, I think cloud players fit in very naturally into this equation, but I think what also has to happen is just like in, I love your word of intentional innovation, ? What they also have to write. I think they have to take a little bit of intentional approach in connecting. thier are software components to the low code world, much more intentionally. I'll explain to you what it is, ? The cloud providers are very much. In the past they've been they've been providing infrastructure components, like network bandwidth, CPU service, but lately they've thought a become the providers of pickaxes and shovels. ? Think about home Depot. You walk into home Depot, [00:26:00] racks, and racks and racks and shelves of, Two by fours and tools and, drillers and nails and screws and everything.

You can find every component, but they're very granular component, ? The reason they provide that is because they want to be, they want to be the building blocks to every company out in the world so they can take advantage of volume and scale and empower the development. It goes back to a little bit of, they want to be the.

These walled gardens or these all gateway, depending on how you look at each one of them off component software, they essentially, they want to be part B. This provider is one stock providers of component software, ? So that's the paradigm where they're all going to, but the next evolution of component software is the dock where all the software is.

And then not only you have, you are putting together software in these docs, but you're also putting together your, you are reselling and reusing software, existing ships that you already do. So low code becomes that doc where, all these components, software offered by these cloud [00:27:00] providers can be assembled together and build software.

And also it becomes a place where you can resell where ISV and others can resell existing, assembled paradigms, again, in a marketplace again, within their cloud capability. But I think what's also interesting is for the cloud providers, they should look at how they can Do this in a very effective manner, provide these docs to their customers.

Because a lot of the, I think the new war in the cloud space is not fought with, fought between fought in Silicon valley. What I mean is, the Amazons, the Google do the Googles and the Microsoft AWS Azure's of the world have already had their footprint in Silicon valley.

All in LA, like very tech center. Company so that not only Silicon valley, anywhere Austin, anywhere in the world, but very tech, heavy companies, ? The next gen, the next shift of the paradigm or their next market share is going to be coming from traditional organization adopting technology.

But what's really interesting in these traditional organization is number [00:28:00] one. They don't have a very modern five forces of innovation workforce. They have a lot of traditional web engineers. And can you provide a dock where they can easily upscale and go to market rapid? So I think that's why they have to think about the low code and no code especially, low code more significant.

No code is interesting, but I think no code is a little bit of counter intuitive to their paradigm too, because they are essentially encouraging engineers and developers to build no code. Of course is a bit of know, is more meant for marketing and other folks because it's not fully convinced.

So I think that's why they look low code platforms and low code docs are very critical for for cloud players. And they are the thing that I think that's where the next war is going to be for in capturing traditional.

Jeff: Very interesting, Brian earlier in the conversation and I wrote it down.

I meant to get back to it a little bit earlier, but you were talking about self building or self writing software. You want to tell me a little more about that? That actually sounds a little scary, [00:29:00]but I'm sure you're gonna, you're gonna swage by my nerves about that. What is that?

Brian: So essentially what happens is with machine learning, with supervised learning. One of the things that's really interesting is you feed the machine. With a certain set of patterns, and the machine would automatically understand and learn your pattern. And then when you give it a new set, it would basically repeat . That pattern. I could put you into a bucket and understand which pattern it was by using methods like that.

What can happen is, and it has come long ways. How do you actually use machines to look. Code that's been prebuilt. You can have machines learn code that has already been built, applications that are being built and you can have the machine regenerate given the use cases and also have the machine teach the machine.

How things are being broken down. So you could basically go in, it could manifest in many different ways. So one of the simplest manifestation, a lot of companies are trying to do is basically imagine [00:30:00] you log into a portal, ? Or into some dashboard. And you can say, you can select, I want this application, you type a paragraph, say I'm looking for a loyalty application where.

Loyalty points are tracked by the system and consumers can spend and loyalty points can be transferred, blah, blah, blah. There is an interface. There's a dashboard, blah, blah, blah. You type a paragraph or mega paragraph required. And say, click on a button and the system could generate code and understand the intent of every sentence you said, and actually look at it and generate code.

So I think we are a little bit away from that because the software systems are very complex. I've seen companies taking a stab at that. There are startups that actually where you can go and type a paragraph and it would. Parts of it, the code. But I think we are far away from that, but we are not too far away.

But when it can be done, but I think with one piece where it's much more useful is where, a designer can go into a design platform and they can go do a design of these screens of how things have to look at. And they have [00:31:00] the ability to tag what functionality they need. And the backend for that is, and then the front end, a corresponding front end is automatically generated based on all that.

And then it allows to connect to the appropriate backends, ? Because with low code, what's really interesting is you're already componentizing everything, it's already component. And then the low code platform or the low code builder is the dry dock where all these components are there and they can be connected and hooked together.

Like a Lego. So if you teach enough Lego drawings here, . A hundred and like all the other 4 million Lego drawings that are out there, the machine can learn that and putting my generator, the next Lego, Lego picture for you, or the architecture, that's essentially what's happening.

But then low code is making machine learning even better. And then that can regenerate code and then suddenly a lot of the code in low code can be regenerated by. So that's, I think is going to be very interesting, ? And that'll help companies get to market a lot faster, ? Because low code today, based on the measurements we've done can give you a speeds from 10 to 17 X faster, go to markets.

If you are a [00:32:00] business leader, head of marketing, head of innovation, head of digital, head of VP or CTO, VP. Some business unit, you are looking at night, you don't, you care about how quickly can I get something to market that my consumers will be loved, will love. And they would really continue to use, ?

So this low code gives you a paradigm where you can go to market faster, but with low code powered by AI, where it can automatically generate itself, it takes it like, another order of magnitude faster.

Jeff: Wow. That's that's really interesting. I, when I think about what you just described and most of these pieces that we, that you talked about are already here.

It's not like we're projecting out in the future. How will all this influence traditional organizations or what do retailers need to be doing today? To prepare for this. Cause I don't think we're talking about a ten-year adoption pattern. I think we're talking about far faster adoption, better than that.

So what do traditional retailers and traditional organizations that you'd be doing today to take advantage of all. [00:33:00]

Brian: See, I think a couple of things, one is of course read up on a lot of low code because knowledge is power, ? Today, if if you go to our website, iterate.ai you can see a lot of use cases on local, ?

It's not, it's true for every software vendor who is providing low code capabilities. You can read out. And if in the, even if you just Google, retail applications powered by low code ? So today, we've seen, we can power things like recommendation, commerce engines, buy online pickup in stores, license, plate recognition scan and go, endless aisle POS capability. Community, social selling communities, all these things can be powered purely through low code so the thing is to learn a lot as much as possible, ? The other is, top experts in this industry in low code

the other is pick a few use cases because I think what happens is sometimes when something is really new and then, there is this natural human tendency to get over it. And try to do the perfect thing. Don't wait for the perfect thing, like in a. And Amazon is of course, talked a lot in everywhere.

But I think one thing I really like about the paradigm is I think it was based on this, I think [00:34:00]said previously, Decisions based on 70% of the information, ? Pick a few use cases. And then just and especially the ones that are low hanging, ? Because one of the problems in this, the thing I think interesting Jeff is traditional companies always have a problem of identifying what an exact MVP minimal viable product is.

The moment you give an idea of a minimal viable product within three meetings, minimum viable product will become a full featured. Because it's very hard for leaders and, players in the industry to say, okay, I'm going to do a bad board skeleton. And then the doubt of what's with the bad ones, kill it and nobody's going to use it.

But I think, figuring out how to work through those temptations and pick up, pick a few use cases and then build a few things very quickly with low code a minimal viable product, test the market out. And one thing that's really interesting, it's a great deal as a waiter, as a very good with testing, ?

Like it doesn't, I the online paradigm is all centered around AB . Test a lot, pick something small test. Keep iterating, keep adding, and then you get a lot going, and then before, we've [00:35:00] had customers who we started with one application.

Now we have 14 plus application running inside them. Matt, within a matter of year. So I think it's totally possible you can access but the idea is to like, learn pick a few use cases and then and then do the smallest thing fast as possible and then integrate.

Jeff: Very interesting.

So you're really not advocating for a rip and replace. You're advocating for a very small scale project to to practice on. I agree. Wow. Interesting. It sounds like we really need to do a conversation around how a retailer needs to have an innovative DNA. That sounds like maybe that's a conversation for the for another day, Brian, as always very interesting to hear about low code.

I remember years ago when you first started pitching me on low code, I thought you were crazy and yeah, you're crazy like a Fox. So thanks again for all the fall, the good information. And we will continue the conversation.

Brian: That's all the ways. Jeff. Thank you. It's awesome. Talking to you. And I think it has to be out in the Casper [00:36:00] evolution already in the middle of it.

I think, we just come and see more.

Jeff: Alrighty. See you later.

Brian: Take care.