Episode Summary
Executive Summary: The episode ranges over year-end reflections on 2018 tech, criticism of Medium, an explanation of a Google/Edge video hack, ongoing frustrations with Apple’s butterfly keyboards, iPad Pro screen protectors and device bends, Siri/AI leadership changes, and listener Q&A on podcast cookies, Intel vs ARM, editing workflow, and Xcode window management. The tone is skeptical, practical, and highly anecdotal, with repeated emphasis on reliability, ergonomics, and Apple’s design tradeoffs.
Main Topics: 2018 tech reflections and recency bias: The hosts reject a true year-end roundup because it requires too much homework and memory, instead discussing the 'impression' of 2018: late-year Mac issues and a year of Facebook privacy scandals that mostly changed no minds. Medium’s degraded reading/writing experience: A long rant argues Medium’s interstitials, fixed banners, and shifting business models make it bad for readers and poor for creators who want to build their own audience. Google’s Edge video hack may have been non-nefarious: They discuss a case where a transparent div over video on Edge/IE may have been an intentional workaround for keyboard shortcuts rather than an anti-competitive move. Butterfly keyboard reliability problems: The 2018 MacBook keyboard still appears to suffer from double-spacing and mechanical failures despite the membrane redesign, and the hosts explain debouncing and why spacebars are particularly fragile. iPad Pro usability: fingerprints, screen protectors, and bends: Marco describes using a matte Paperlike protector to reduce fingerprints and improve pencil feel, while the conversation also covers Apple acknowledging small bend tolerances and the difficulty of judging whether an iPad is defective. Apple’s Siri/AI leadership and product quality: John Giannandrea’s promotion is treated as a positive signal, but both hosts want visible product improvements, especially in Siri speed, reliability, and cross-device behavior. Ask ATP: cookies, user agents, Intel/ARM, editing, and Xcode full screen: Listener questions prompt explanations of ephemeral cookie handling in Overcast, why provider blocking is unlikely, why Apple may still move to ARM, how ATP editing works, and why full-screen Xcode is unpopular for most of them.
Key Arguments: Year-end tech retrospectives are hard because important early-year stories fade from memory and recency bias dominates impressions. Medium is useful only as a distribution venue; it is poor for creators who want ownership, audience-building, or a clean reading experience. The Google/Edge video issue may have been caused by a legitimate website workaround for browser keyboard shortcuts, not necessarily anti-competitive sabotage. Butterfly keyboards remain fundamentally fragile; the membrane improves things only marginally because the underlying design is still problematic. The spacebar is especially failure-prone because of its size and mechanism, making double-spacing a plausible mechanical defect rather than a software debounce issue. Apple’s iPad Pro bend tolerance discussion was mishandled in PR; even if a tiny bend is within spec, visibly bent units should likely be exchanged. Siri’s quality problems are not just about intelligence but also speed, reliability, and the willingness to execute requests without deferring to the phone. Apple’s AI leadership elevation matters less than whether products become materially better. Podcast apps do not need long-term cookie storage to download episodes; ephemeral sessions can use cookies transiently without saving them. Apple switching to ARM still feels likely because control and in-house economics matter more than Intel’s ability to recover later. Full-screen mode is widely disliked by these hosts for development because multi-window workflows are more flexible and less disruptive.
Data Points: 2018 recap effort: Too much work / not done - Hosts say year-end roundup episodes require more prep than they’re willing to do. Butterfly keyboard generation: 2018 - The discussion centers on the 2018 MacBook Air keyboard and its spacebar issues. Spacebar failure symptom: Double spaces - Reported as a common problem on 2018 butterfly keyboards. Debouncing timing reference: 16-bit, 8-megahertz processor - Used as an example of how little processing power is needed to debounce keyboard input. Screen protector brand: Paperlike - Marco describes using the matte iPad screen protector. Screen protector effect: Less fingerprinty - Marco says the iPad still fingerprints, but noticeably less with the protector. Apple bend tolerance: 400 microns - Dan Riccio reportedly described some amount of bend as normal up to this amount. Paper comparison for bend: 4 sheets of paper - Used to explain the scale of 400 microns. Siri improvement claim: ~20% better than last year - Referenced from a recent assistant comparison test. Backblaze pricing: $5 per month per computer - Promotional and descriptive pricing for unlimited cloud backup. Away promo: $20 off - Discount offered with promo code ATP. MacWeldon promo: 20% off first order - Discount offered with promo code ATP. CrashPlan memory allocation: 5 GB - Marco describes needing to assign five gigabytes of RAM for CrashPlan to function. CrashPlan plan cost increase: $5 to $15 per month - Described as an increase after a discount expired. Backblaze B2 backup cost: about $12/month - Marco’s planned Synology backup replacement.
Pivotal Quotes: "I friggin' hate Medium so hard." — John: A rant about Medium’s login prompts, banners, and poor reading experience. "This design is problematic. It's just a little bit better protected against its own horrendous design." — John: Explaining why 2018 butterfly keyboards may still fail despite the membrane. "If you get one that's bent, bring it back and we'll exchange it." — John: What Apple should have communicated about bent iPad Pros instead of saying bend is normal.
Implications: The episode reinforces a recurring theme: polished hardware and smart services still fail when small usability and reliability details are wrong. Listeners are likely to be skeptical of platform promises and value practical workarounds over hype.