Scott & Mark Learn To...

In this episode of Scott and Mark Learn To, Scott Hanselman and Mark Russinovich delve into the intricate world of undocumented APIs, examining their associated risks and benefits. Mark shares his journey into reverse engineering, starting with the Apple II and evolving through his exploration of Windows internals. He elaborates on his PhD thesis on fault management, which sparked his interest in undocumented APIs. The conversation includes technical insights into disassembling and debugging, showcasing Mark’s development of his own disassembler and use of tools like SoftICE for kernel debugging. They also explore Microsoft’s rationale for keeping certain APIs undocumented, highlighting the importance of flexibility in system evolution without compromising application stability.Takeaways: Microsoft’s commitment to maintaining backward compatibility for older applications How contemporary systems manage small crashes to maintain long-term stability How Mark started his journey on Windows internal APIs Who are they? View Scott Hanselman on LinkedIn View Mark Russinovich on LinkedIn Watch Scott and Mark Learn To on YouTube Listen & Subscribe at scottandmarklearn.to Discover and follow other Microsoft podcasts at microsoft.com/podcastsDownload the Transcript Hosted on Acast. See acast.com/privacy for more information.

In this bonus episode of Scott & Mark Learn To, Scott Hanselman and Mark Russinovich take a detour into the fabulous world of skincare, proving that everyone deserves a little TLC! With Scott both amused and bewildered by Mark's radiant complexion, they launch into a lively debate about grooming habits. From the champagne of shampoos to moisturizing rituals and the must-have essentials for looking sharp in the tech world, Scott and Mark share their best beauty tips. Plus, discover the profound truth that “you can’t have too many moisturizers”—unless you're Mark, who insists that sometimes, less is more! Takeaways: Mark's glowing skin leads Scott to question how old he really is The importance of skincare routines as a form of self-care A minimalistic routine vs a more extensive regimen and common misconceptions Who are they? View Scott Hanselman on LinkedIn View Mark Russinovich on LinkedIn Watch Scott and Mark Learn To on YouTube Listen & Subscribe at scottandmarklearn.to Discover and follow other Microsoft podcasts at microsoft.com/podcasts Download the Transcript Hosted on Acast. See acast.com/privacy for more information.

In this episode of Scott & Mark Learn To, Scott Hanselman and Mark Russinovich dive into the quirks and philosophies of programming languages—debating the purpose of small languages, like Erlang, versus giants like JavaScript. Along the way, they discuss whether modern languages with hindsight (like Kotlin) are inherently better and reminisce about the good old days of writing code in C, Delphi, and even 6502 assembler. Mark reflects on transitioning to Rust for system-level code in Azure, marking a deliberate move away from C/C++. He highlights Rust’s ownership model, memory safety, and enforced concurrency as the game-changers. Meanwhile, Scott is poking fun at the oddities of language trends, including Python’s rise to dominance in machine learning. Takeaways: Rust’s key advantage with enforced memory safety Python’s dominance in AI due to its robust ecosystem not just its language features Why programming language selection is driven by ecosystems, tools, and personal preference Who are they? View Scott Hanselman on LinkedIn View Mark Russinovich on LinkedIn Watch Scott and Mark Learn on YouTube Listen to other episodes at scottandmarklearn.to Discover and follow other Microsoft podcasts at microsoft.com/podcasts Download the Transcript Hosted on Acast. See acast.com/privacy for more information.

In this episode of Scott & Mark Learn To, Scott Hanselman and Mark Russinovich explore the evolving role of AI in tech, from leveraging tools like GitHub Copilot to boost productivity in coding, to the potential pitfalls of over-reliance on AI. They discuss how AI is reshaping both education and professional development and reflect on the challenges of large language models (LLMs), including issues like hallucinations, indirect prompt injection attacks, and jailbreaks. Mark highlights how models, shaped by Reinforcement Learning with Human Feedback (RLHF), can still produce unpredictable results, underscoring the need for transparency, safety, and ethical use in AI-driven systems. Takeaways: Whether reliance on AI affects one's foundational coding skills and overall efficiency How to balance continuous learning with maintaining expertise in technology fields AI models sometimes produce hallucinations and the importance of understanding how to effectively use these tools Who are they? View Scott Hanselman on LinkedIn View Mark Russinovich on LinkedIn Listen to other episodes at scottandmarklearn.to Watch Scott and Mark Learn on YouTube Discover and follow other Microsoft podcasts at microsoft.com/podcasts Download the Transcript Hosted on Acast. See acast.com/privacy for more information.