Avsnitt
-
How do you balance the coupling in your application? Carl and Richard talk to Vlad Khononov about his book on Balancing Coupling in Software Design. Vlad talks about three aspects of coupling - information, distance, and volatility. When these aspects are out of balance, such as a pair of services that are distant from each other but highly dependent and need lots of information, development becomes difficult. Where information is high, keeping the distance low makes life easier. This led to a great conversation about Conway's Law and the idea that sometimes changing the team organization can lead to better application development! Check out the book!
-
What's coming for Blazor in .NET 9? Carl and Richard talk to Dan Roth about the upcoming version of Blazor. Dan discusses excellent performance improvements, better MAUI interactions, new SignalR features, and more! The conversation also dives into how Blazor gets made and the journey that submitting issues into GitHub goes through to become features in the Blazor framework. It takes a while, but you can be part of making Blazor great!
-
Saknas det avsnitt?
-
What does it mean to build cloud-native applications? Carl and Richard talk to Chris Klug about his experiences building applications designed to operate effectively in the cloud. Chris pushes back on the fixation around Kubernetes - you can build cloud-native apps without it! The conversation digs into the various options available to take advantage of the cloud's ability to scale while also tolerating its occasional short-duration outages and shifting availability. Chris also talks about .NET Aspire and its ability to help you build .NET cloud-native applications.
-
What has Chris Sells been up to? Carl and Richard chat with Chris Sells, the guest on episode 10 back in 2002, about how his career continues to evolve. Chris talks about working at Google on Flutter, the mobile dev stack - before departing for Meta to work on the tooling for augmented reality. The conversation digs into how AR appears to be the logical evolution of mobile but has been completely overwhelmed by artificial intelligence. Chris has left Meta to work on AI technologies and sees huge potential in making better applications than ever before!
-
How do you make your own copilot? Carl and Richard talk to Prashant Bhoyar about his work with Copilot Studio and Azure AI Studio. Prashant describes how Copilot Studio lives in the Power Platform space while Azure AI Studio is more related to Visual Studio, in that it is a tool for developers of AI technology. Anything built in Azure AI Studio can be surfaced in Copilot Studio - another kind of fusion development! Lots of conversation about what works well and what is difficult with these tools, and how to avoid some critical mistakes!
-
Ready for a chat with the creator of Node? Carl and Richard talk to Ryan Dahl about his work creating NodeJS in 2009 and how he moved on after a few years, leading to the creation of Deno, an opinionated approach to building web applications. Ryan talks about the challenges of simplifying web development by combining all the important things into a single set of tools—saving you the effort of assembling those things yourself. The conversation also digs into how web development has evolved and one of Ryan's current efforts - convincing Oracle to surrender the JavaScript trademark to the world!
-
What's the latest with Playwright? Carl and Richard talk to Debbie O'Brien about her ongoing work with Playwright, Microsoft's open-source testing framework for web applications. While it is focused on web applications, you can write your tests in various languages, including .NET! Debbie talks about the new Playwright Testing service, which operates in Azure, so you don't have to stand up with your testing infrastructure - pay for what you use. The conversation ranges over the various features and challenges in testing that Playwright addresses. Now get out there and write some tests - your applications will be better for it!
-
How simple can you make software development? Carl and Richard talk to Mark Rendle about his focus on simplicity in building software - as simple as possible. Mark talks about the tendency of developers, sometimes through no fault of their own, to use what is new and cool in development, regardless of how practical or necessary it actually is. The conversation digs into the ongoing battle around cloud-native development using technologies like Kubernetes. You can be cloud native with more straightforward approaches! The same applies to web frameworks - there are lots of choices. Build as little as necessary!
-
How does good design pay off in software? Carl and Richard talk to Billy Hollis about his work designing software, both from a user interaction perspective and application architecture. Billy talks about saving time and money by working hard on design to get a clearer picture of what stakeholders want—because code rework is always more expensive! The conversation also digs into the institutional knowledge walking out of many companies through employees retiring—and how much work that is going to generate over the next few years to modernize!
-
Can you do mob programming remotely? Carl and Richard talk to Ulrika Malmgren about mob, or team programming - where three to five developers work together on the same problem, rotating keyboard control and collaborating to write the best possible code. Modern tools like Teams and Zoom make it easy to do the same programming style with everyone remotely! Ulrika talks about how team programming becomes the standard approach to development - as opposed to solo programming, where everyone works separately and then has to spend time sharing what they've done! Good team programming results in higher-quality code in less time - and more happiness!
-
What can machine learning do for geospatial data? Carl and Richard talk to Malte Loller-Anderson and Mathilde Ørstavik about their work at Norkart, using aerial imagery to build detailed maps around Norway. Mathilde dives into the critical role of machine learning - identifying buildings in images. Usually done by hand with each new image, Norkart has a machine learning model that automates the process trained on previous vector maps of buildings. But there are many things that look like buildings in Norway, including patches of snow, mountains, and even shapes under water. Malte also discusses how Norkart has decided to train in-house with nVidia L40 processors rather than in the cloud - the hardware is used 24 hours a day since some models can take weeks to train! There are many interesting ideas about geospatial data and machine learning from people who have been doing it for years.
-
What do Domain-Driven Design and event sourcing have to do with each other? Everything! Carl and Richard chat with Anita Kvamme about her experiences applying DDD, and specifically event storming, to developing applications using event sourcing. Anita talks about building applications that have many sources of events—from users and elsewhere—and needing to manage that complexity without slowing down development. Event sourcing also means keeping a source of the truth - all events leading up to a practical business benefit. And that can be hugely helpful in analytics as well!
-
How can a low-code solution help you deliver a .NET app? Carl and Richard talk with Serge Sarafudinov about his Xomega project. Serge describes how Xomega uses models and templates to generate .NET code for applications for Blazor clients, WPF, and even ASP.NET Forms and TypeScript! The conversation also digs into rehabilitating existing .NET applications where new features can be added with Xomega, and then gradually convert the existing application into the model approach - and then you can change out the client if you like! There are free and paid versions of Xomega; take it out for a spin and see if you can't deliver solutions faster!
-
How do you manage APIs to GenAI, and how can GenAI help with API management? Carl and Richard chat with Andrei Kamenev about the latest features coming to Azure API Management. On the one hand, there are Copilot tools to help craft and understand APIM policies, which can get very complex. Then, there is the provisioning of access to GenAI-related APIs like the Azure OpenAI service, which utilize tokens - and those tokens mean money, so they need to be controlled. The GenAI Gateway provides the ability to rate-limit token issuing and all the other capabilities you expect from APIM. Prompt caching is in preview and can decrease the cost of repeated use of the same prompts. Many of the features are new, and more are coming!
-
How can .NET Aspire help you get into containers? Carl and Richard talk with Jiachen Jiang about her experiences working with .NET Aspire. Jiachen talks about the power of .NET Aspire to help lead developers to utilize cloud native architecture efficiently. A key part of the equation is containers, and Jiachen talks about how relatively rare containerized .NET apps are - because in many cases, it doesn't add anything but complexity. But as .NET applications need to scale in the cloud, containers become important, and .NET Aspire helps you to take advantage of containers while adding all the cloud-native features you'll value, like great telemetry, orchestration, and discovery!
-
What if you want to build your own copilot? Carl and Richard talk to Vishwas Lele about his new startup, which is focused on using Azure OpenAI tools to help automate the government RFP writing process. Vishwas discusses the complexities of proposal writing, how specific and complex rules exist for each part of the proposal, and the challenge of getting the software to do an excellent job on the draft. The conversation digs into the domain expertise needed for the technologies and the proposal writing itself - like all good software, it requires domain experts. But when done right, this is hugely valuable software!
-
What are the basics of building a decent user interface on a web page? Carl and Richard talk to Hilary Stohs-Krause about her experience helping folks build websites - you don't have to be a designer to make something useable! Hilary talks about steering clear of more complex UX libraries unless you know the site owner prefers them. Keeping things simple and consistent is best! The conversation also digs into accessibility and how accessibility needs are a great guide to a decent basic design - making a site that can help everyone understand the pages you've built a great path to a very useable website.
-
What's happening with WPF? Carl and Richard talk to Joseph Finney about the news from Build about WPF - it's back, baby! But did it ever go away? Joe talks about how WPF was made open source in 2018 and how minimal development was done there - enough so that the community started building frameworks to support it, including WPFUI and ModernWPF. But at Build, Microsoft announced a new focus on WPF, including creating support for a Windows 11 theme - with a line of code, you can make your WPF app look like a Windows 11 app. Where does this leave WinApp SDK? What about migrations? There's lots to talk about!
-
Multi-modal is here and ready to use! Carl and Richard talk to Veronika Kolesnikova about what she has seen in the latest OpenAI model. Multi-modal refers to GPT-4o's ability to work with text, audio, images, and more and respond in kind! Veronika talks about tackling interesting visual problems with GPT-4o - like summarizing graphs or creating new data visualizations. The conversation digs into some announcements around Build, including the new security features in AI Studio and Windows Recall in the new Copilot+ PC devices. Things are moving quickly!
-
CosmosDB makes ChatGPT fast! While at Build in Seattle, Carl and Richard chatted with Mark Brown about CosmosDB's role in AI. Mark talks about how ChatGPT switched over to CosmosDB early on - when the number of users started to climb, database performance became essential, and CosmosDB was there. Today, many AI-centric CosmosDB features exist, like vector storage, indexing, and search! The conversation also digs into the impact of the large language model on development - things are different now!
- Visa fler
