The New Stack Podcast

Nearly seven years after Google released Kubernetes, the open source container orchestrator, into an unsuspecting world, 5.6 million developers worldwide use it.But that number, from the latest Cloud Native Computing Foundation (CNCF) annual survey, masks a lot of frustration. Kubernetes (K8s) can make life easier for the organization that adopts it — after it makes it a lot harder. And as it scales, it can create an unending cadence of triumph and challenge.In other words: It’s complicated.At KubeCon + CloudNativeCon EU in Valencia, Spain last week, a trio of experts — Saad Malik, chief technology officer and co-founder of Spectro Cloud; Bailey Hayes, principal software engineer at SingleStore; and Fabrizio Pandini, a staff engineer at VMware — joined Alex Williams, founder and publisher of The New Stack, and myself for a livestream event.

The pandemic has significantly accelerated the adoption of Kubernetes and cloud native environments as a way to accommodate the surge in remote workers and other infrastructure constraints. Following the beginning of the pandemic, however, organizations are retaining their investments for those organizations with cloud native infrastructure already in place. They have realized that cloud native is well worth maintaining their investments. Meanwhile, Kubernetes adoption continues to remain on an upward curve. And yet, challenges remain, needless to say. In this context, we look at the status of cloud native adoption, and in particular, Kubernetes at this time, compared to a year ago. In this episode of The New Stack Makers podcast, recorded on the floor of KubeCon + CloudNativeCon Europe 2022, we discussed these themes along with the state of Kubernetes and the community with James Laverack, staff solutions engineer, Jetstack a member of the Kubernetes release team, and Christoph Blecker, site reliability engineer, Red Hat, a member of the Kubernetes steering committee.

 Go was created at Google in 2007 to improve programming productivity in an era of multi-core networked machines and large codebases. Since then, engineering teams across Google, as well as across the industry, have adopted Go to build products and services at massive scale, including the Cloud Native Computing Foundation which has over 75% of the projects written in the language.In this episode of The New Stack Makers podcast, Steve Francia, Head of Product: Go Language, Google and alumni of MongoDB, Docker and Drupal board member discusses the programming language, the new features in Go 1.18 and why Go is continuing on a path of accelerated adoption with developers. Darryl Taft, News Editor of The New Stack hosted this podcast.In the State of Developer Ecosystem 2021, Go ranked in the top five languages that developers planned to adopt and continues to be one of the fastest growing languages. According to Francia, it was created with the motivation to see if a new system programming language could be built and compile quick with security as the top focus. With developers coming and going at Google, the simplicity and scalability of the language enabled many to contribute across several projects at any given time.“The influences that separates Go from most languages is the experience of the creators behind it who all came to build it with their collective experience,” Francia said.  Today “Go is influencing a lot of the mainstream languages. Elements of it can be found in a tool that formats everyone’s source code to be identical and more readable. Since then, a lot of languages have adopted that same practice,” said Francia. “And then there’s rust. Go and rust are on parallel tracks and we're learning from each other. There's also a new language called V that has recently been open sourced which is the first major language inspired by Go,” Francia said.The latest release of Go 1.18 was Google’s biggest yet. “It included four major features, each of which you could build a release around,” said Francia. In this release, “Generics is the biggest change of the Go language which has been in the works for 10 years,” Francia added. “Because we knew that generics have the potential to make a language more complicated, we spent a long time going through different proposals,” he said. Fuzzing, workspaces and performance were three other features released in this past version of Go.“From improving our documentation and learning – which you can go to go.dev/learn/ to get the latest resources – we’re really focused on the broad view of the developer experience,” Francia said. “And in the future, we're seeing not our team so much as the community taking Go in new ways,” he added.    

First released in 2016, the Svelte Web framework has steadily gained popularity as an alternative approach to building Web applications, one that prides itself on being more intuitive (and less verbose) than the current framework du jour, Facebook's React. You can say that it reaches back to the era before the web app — when desktop and server applications were compiled — to make the web app easier to develop and more enjoyable to user. In this latest episode of The New Stack Makers podcast, we interview the creator of Svelte himself, Rich Harris. Harris started out not as a web developer, but as a journalist who created the framework to do immersive web journalism. So we were interested in that. In addition to delving into history, we also discussed the current landscape of Web frameworks, the Web's Document Object Model, the way React.js updates variables, the value of TypeScript, and the importance SvelteKit. We also chatted about why Vercel, where Harris now works maintaining Svelte, wants to make a home for Svelte. TNS Editor Joab Jackson hosted this conversation. Below are a few excerpts from our conversation, edited for brevity and clarity. So set the stage for us. What was the point that inspired you to create Svelte? To fully tell the story, we need to go way back into the mists of time, back to when I started programming. My background is in journalism. And about a decade ago, I was working in a newsroom at a financial publication in London. I was very inspired by some of the interactive journalism that was being produced at places like the New York Times, but also the BBC and the Guardian and lots of other news organizations, where they were using Flash and increasingly JavaScript, to tell these data rich interactive stories that couldn't really be done any other way. And to me, this felt like the future of journalism, it's something that was using the full power of the web platform as a storytelling medium in a way that just hadn't been done before. And I was very excited about all that, and I wanted a piece of it. So I started learning JavaScript with the help of the help of some some friends, and discovered that it's really difficult. Particularly if you're doing things that have a lot of interactivity. If you're managing lots of state that can be updated in lots of different ways, you end up writing what is often referred to as spaghetti code. And so I started building a toolkit, really, for myself. And this was a project called Reactive, short for interactive, something out of a out of a Neal Stephenson book, in fact, and it actually got a little bit of traction, not it was never huge, but you know, it was my first foray into open source, and it got used in a few different places. And I maintained that for some years, and eventually, I left that company and joined the Guardian in the U.K. And we used Reactive to build interactive pieces of journalism there, I transferred to the U.S. to continue at the guardian in New York. And we use directive quite heavily there as well. After a while, though, it became apparent that, you know, as with many frameworks of that era, it had certain flaws. A lot of these frameworks were built for an era in which desktop computing was prevalent. And we were now in firmly in this age of mobile, first, web development. And these frameworks weren't really up to the task, primarily because they were just too big, they were too big, and they were too bulky and they were too slow. And so in 2016, I started working on what was essentially a successor to that project. And we chose the name Svelte because it has all the right connotations. It's elegant, it's sophisticated. And the idea was to basically provide the same kind of development experience that people were used to, but change the was that translated into the experience end users have when they run it in the browser. It did this by adopting techniques from the compiler world. The code that you write doesn't need to be the code that actually runs in the browser. Svelte was really one of the first frameworks to lean into the compiler paradigm. And as a result, we were able to do things with much less JavaScript, and in a way that was much more performant, which is very important if you're producing these kinds of interactive stories that typically involve like a lot of data, a lot of animation Can you talk a bit about more about the compiler aspect? How does that work with a web application or web page? So, you know, browsers run JavaScript. And like nowadays, they can run WASM, too. But JavaScript is the language that you need to write stuff in if you want to have interactivity on a web page. But that doesn't mean that you need to write JavaScript, if you can design a language that allows you to describe user interfaces in a more natural way, then the compiler could turn that intention into the code that actually runs. And so you get all the benefits of declarative programming but without the drawbacks that historically have accompanied that. There is this trade off that historically existed: the developer wants to write this nice, state driven declarative code and the user doesn't want to have to wait for this bulky JavaScript framework to load over the wire. And then to do all of this extra work to translate your declarative intentions into what actually happens within the browser. And the compiler approach basically allows you to, to square that circle, it means that you get the best of both worlds you're maximizing the developer experience without compromising on developer experience. Stupid question: As a developer, if I'm writing JavaScript code, at least initially, how do I compile it? So pretty much every web app has a build step. It is possible to write web applications that do not involve a build step, you can just write JavaScript, and you can write HTML, and you can import the JavaScript into the HTML and you've got a web app. But that approach, it really doesn't scale, much as some people will try and convince you otherwise. At some point, you're going to have to have a build step so that you can use libraries that you've installed from NPM, so that you can use things like TypeScript to optimize your JavaScript. And so Svelte fits into your existing build step. And so if you have your components that are written in Svelte files, it's literally a .SVELTE extension. Then during the build step, those components will get transformed into JavaScript files. Svelte seemed to take off right around the time we heard complaints about Angular.js. Did the frustrations around Angular help the adoption of Svelte? Svelte hasn't been a replacement for Angular because Angular is a full featured framework. It wants to own the entirety of your web application, whereas Svelte is really just a component framework. So on the spectrum, you have things that are very focused on individual components like React and Vue.js and Svelte. And then at the other end of the spectrum, you have frameworks like Angular, and Ember. And historically, you had to do the work of taking your component framework and figuring out how to build the rest of the application unless you were using one of these full-featured frameworks. Nowadays, that's less true because we have things like Next.js, and remix-vue, And on the Svelte team are currently working on SvelteKit, which is the answer to that question of how do I actually build an app with this? I would attribute the growth in popularity is felt to different forces. Essentially, what happened is it trundled along with a small but dedicated user base for a few years. And then in 2019, we released version three of the framework, which really rethought the authoring experience, the syntax that you use to write components and, and the APIs that are available. Around that time, I gave a couple of conference talks around it. And that's when it really started to pick up steam. Now, of course, we're growing very rapidly. And we're consistently at the top of developer-happiness surveys. And so now, like a lot of people are aware of is, but we're still like a very tiny framework, compared to the big dogs like React and Vue. You have said that part of the Svelte mission has been to make web development fun. What are some of Svelte's attributes that make it less aggravating for the developer? The first thing is that you can write a lot less code. If you're using Svelte, then you can express the same concepts with typically about 40% less code. There's just a lot less ceremony, a lot less boilerplate. We're not constrained by JavaScript. For example, the way that you use state inside a component with React, you have to use hooks. And there's this slightly idiosyncratic way of declaring a local piece of state inside the component. With Svelte, you just declare a variable. And if you assign a new value to that variable, or if it's an object, and you mutate that object, then the compiler interprets that as a sign that it needs to update the component.  

First released in 1995, Java’s programming language has been a leading developer platform that has become a workhorse for hundreds of enterprise applications. With each new technology evolution, Java has successfully adapted to change. But even while a recent Java ecosystem study found that more than 70% of Java applications in production environments are running inside a container, there continues to be hurdles the language must overcome to adapt to the cloud-native world.In this episode of The New Stack Makers podcast, Simon Ritter, deputy CTO of Azul Systems and Dalia Abo Sheasha, Java developer advocate of JetBrains discuss some of the challenges the language is working to overcome, and share some insight into the new features that developers are requesting. Darryl Taft, news editor of The New Stack hosted this podcast.The complexity of modern applications requires developers to master a growing array of skills, technologies, and concepts to develop in the cloud. And “what I've seen is that there is a gap in skills, and what it would take to get existing Java applications into the cloud,” said Abo Sheasha.“What developers really want is to focus on the idea of developing the Java code,” said Ritter. “Having the ability to plug in to different cloud providers, but also the ability to integrate with things like your CI/CD tooling so that you've got continuous integration, continuous deployment built in,” he added. Getting Java ready for the cloud is a “distributed responsibility across the people – from cloud providers to tooling providers,” said Ritter. “Everyone recognizes that the more folks we have on it, the more minds we have on it, the better outcome we're going to have for the developer’s language,” Abo Sheasha said. Making developers more efficient and productive is coming into the fold with the introduction of JEP, or JDK Enhancement Proposals - a lightweight approach to add new features in the development of the Java platform itself. “But there's some bigger projects like Project Amber which is all about small changes to the language syntax of Java with the idea of making it more productive by taking some of the boilerplate code out,” Ritter said. The journey to the next chapter of Java is multi-dimensional. While “most developers are focused on getting the job done, picking up skills for new things is a challenge because it takes time. Many still have the issue of using whichever Java version their company is stuck on,” said Ritter. “It's not because the developers don't want to do it; it’s that they need to convince management that it's worth investing in,” added Abo Sheasha.

Last week, the country of Spain dropped its mandate for residents and visitors to wear masks, to ward off further infections of the Coronavirus. So, for this year's KubeCon + CloudNativeCon  Europe conference, to be held May 16 - 20th of May in Valencia, Spain, the Cloud Native Computing Foundation dropped its own original mandate that attendees wear masks, a rule that had been in place for its other recent conferences.This turned out to be the wrong decision, CNCF admitted a week later. A lot of people who already bought tickets were upset at this laxing of the rules for the conference, which could put them in greater danger of contacting the disease.So the CNCF put the mandate back in place, and offered refunds for those who felt Spain's own decision would put them in harm's way. CNCF will even send you a week's worth of N95 masks if you request them.So, long story short: bring a mask to KubeCon. And, as always, it is still a requirement to show proof of vaccination and temperature checks will be made as well.Tricky business running a conference in this time, no?In this latest episode of The New Stack Makers podcast, we take a look at what to expect from this year's KubeCon EU 2022. Our guests for this podcast are Priyanka Sharma, the executive director of CNCF, and Ricardo Rocha, who is a KubeCon co-chair and computer engineer at CERN. TNS Editor-in-chief Joab Jackson hosted this podcast.We recorded this podcast prior to the discussion around masks, and at the time, Sharma said that the CNCF based the mask ruling on Spain's own country-wide mandates. "So we are being very cautious with the health requirements for the event," she said.The conference team is also keeping an eye on Russia's aggressive moves in the Ukraine, though it is unlikely that the chaos will reach all the way to Spain. Still, "this is why it's essential to always have the hybrid option .. [to] have the virtual elements sorted," Sharma said.As the CNCF flagship conference, KubeCon brings together managers and users of a wide variety of cloud native technologies, including containerd, CoreDNS, Envoy, etcd, Fluentd, Harbor, Helm, Istio, Jaeger, Kubernetes, Linkerd, Open Policy Agent, Prometheus, Rook, Vitess, Argo, CRI-O, Crossplane, dapr, Dragonfly,  Falco, Flagger, Flux, gRPC, KEDA, SPIFFE, SPIRE, and Thanos, and many many more. Most have been featured on TNS at one time or another.In this podcast, we also discuss what to expect from the virtual sessions at the conference, what to do in Valencia, the current state of Kubernetes, and we get some unofficial picks from Sharma and Rocha as to what keynotes not miss and what sessions to attend."The virtual option is great," Rocha said. "But I think the in-person conferences have have their own value. And there's a lot to be to be gained about meeting people directly and exchanging ideas and going to these events on the side of the conference as well."

Low-code and no-code is becoming increasingly popular in software development, particularly in enterprises that are looking to expand the number of people who can create applications for digital transformation efforts. While in 2020, less than 25% of new apps were developed using no code/low code, Gartner predicts that by 2025, 70% will utilize this means. Microsoft is one vendor who has been paving the way in this shift by reducing the burden on those in the lines of business and developers in exchange for speed. But what are the potential and best practices for low code/no code software development? In this episode of The New Stack Makers podcast, Charles Lamanna, Corporate Vice President, Business Apps and Platform at Microsoft discusses what the company is doing in the low-code/no code space with its Power Platform offering, including bringing no code/low-code professionals together to deliver applications. Joab Jackson, Editor-in-Chief of The New Stack and Darryl Taft, News Editor of The New Stack hosted this podcast. 

Developers are often faced with complexity when building and operating long-running processes that involve multiple service calls and require continuous coordination. To solve this challenge, Uber built and introduced Cadence, the open-source solution for workflow orchestration in 2016 that enables developers to directly express complex, long-running business logic as simple code. Since its debut, it continues to find increased traction with developers operating large-scale, microservices-based architectures. More recently, Instaclustr announced support for a hosted version of Cadence.In this episode of The New Stack Makers podcast, Ben Slater, Chief Product Officer at Instaclustr and Emrah Seker, Staff Software Engineer at Uber discuss Cadence, and how it is used by developers to solve various business problems by enabling them to focus on writing code for business logic, without worrying about the complexity of distributed systems.Alex Williams, founder and publisher of The New Stack hosted this podcast, along with co-host Joab Jackson, Editor-in-Chief of The New Stack.

As the tech stack grows, the list of technologies that must be configured in cloud computing environments has grown exponentially and increased the complexity in the IT infrastructure. While every layer of the stack comes with its own implementation of encrypted connectivity, client authentication, authorization and audit, the challenge for developers and DevOps teams to properly set up secure access to hardware, software throughout the organization will continue to grow, making IT environments increasingly vulnerable. In this episode of The New Stack Makers podcast, Ben Arent, Developer Relations Manager, Teleport discusses how to address the hardware, software and peopleware complexity that comes from the cloud by using tools like Teleport 9.0 and the company’s first release of Teleport Machine ID. 

From cloud security providers to open source, trust has become a staple from which an organization's security is built. But with the rise of cloud-native technologies, the new ways of building applications are challenging the traditional approaches to security. The changing cloud-native landscape is requiring broader security coverage across the technology stack and more contextual awareness of the environment. So how should DevOps and InfoSec teams across commercial businesses and governments rethink their security approach? In this episode of The New Stack Makers podcast, Tom Bossert, president of Trinity Cyber (and former Homeland Security Advisor to two Presidents); Patrick Hylant, client executive of VMware; and Chenxi Wang, managing general partner, Rain Capital discuss how businesses and the U.S. government can adapt to the evolving threat landscape, including new initiatives and lessons that can be applied in this high-risk environment. Alex Williams, founder and publisher of The New Stack, hosted this podcast. Jim Douglas, CEO of Armory also joined as co-host of this livestream event.