MobileViews Podcast 573:A Tsunami of tech topics + Uncle Jon’s Bank for Parents & Kids

Todd Ogasawara and Jon Westfall covered a range of interesting topics, from real-world natural disasters to the cutting edge of AI development and personal tech. Todd shared his recent experience during a statewide tsunami alert in Hawaii, triggered by an 8.8 magnitude earthquake off Russia. While initial information was well-managed, he highlighted significant issues with traffic chaos during evacuation and a concerning lack of information post-wave impact. On the technology front, Todd discussed Google Notebook LM, praising its ability to create succinct summaries and slideshows with voiceovers from source material. He also introduced Google Opal, a new experimental tool from Google Labs that allows users to build and share powerful AI mini-apps using natural language and visual editing, describing it as a "step beyond Visual Basic" for accelerating AI prototyping and workflows.

Jon Westfall also shared his recent tech purchases and an exciting new project. He acquired an 8Bitdo Micro Bluetooth Gamepad, a pocket-sized mini-controller weighing just 24.8 grams with 16 buttons. Its versatility allows it to function as a game controller for Switch, Android, and Raspberry Pi, or as a keyboard mode device for various applications, including as a remote for his new Kobo Libra Colour eReader. The Kobo Libra Colour features a 7" E Ink Kaleido™ 3 color display and Kobo Stylus 2 compatibility for colorful mark-ups and note-taking, with notebooks backed up to Kobo Cloud, Dropbox, or Google Drive. Jon also unveiled his open-source project, Uncle John's Bank, a self-hostable banking system for parents and kids designed to teach financial literacy, notably incorporating daily compounding interest and Certificates of Deposit (CDs). This sophisticated project was developed remarkably fast (75 hours) thanks to extensive use of OpenAI Codex, which integrated directly with his GitHub repository, even writing developer documentation. However, Jon noted a peculiar issue where GitHub Copilot (AI) reviewing Codex (AI)-generated code sometimes caused new problems, suggesting limitations in AI-to-AI code interaction. Finally, Jon shared intriguing results from asking various AIs (Google Gemini, ChatGPT, Microsoft Copilot, Anthropic Claude) for investment advice, observing their diverse recommendations and risk appetites.