cbsnews.com
Judge denies Karmelo Anthony's appeal for a new trial in killing of Austin Metcalfhttps://www.cbsnews.com/news/karmelo-anthony-new-trial-decision-austin-metcalf-8-22-2026/The decision comes after both prosecutors and Anthony's new defense team revealed more details in a tense two-day hearing.Aug 23, 2026 4:15 AM

cbc.ca
B.C. family feels 'violated' after burglary during wildfire evacuationhttps://www.cbc.ca/news/canada/british-columbia/wildfire-evacuation-robbery-summerland-9.7317262?cmp=rssMira Boucher says her home was ransacked by looters while her community of Summerland was empty and vulnerable — with every single house vacant under an evacuation order.Aug 23, 2026 2:00 AM

france24.com
Macron hosts Saudi crown prince for two-day visit to include issues from energy to Mideasthttps://www.france24.com/en/europe/20260823-macron-welcomes-saudi-crown-prince-mbs-for-two-day-france-visitSaudi crown prince and de facto ruler Mohammed bin Salman (MBS) was welcomed on Sunday by French President Emmanuel Macron for a two-day visit to the country. Several agreements on health, transport and energy will be signed on the second day of the visit on Monday underlining the strength of the Paris-Riyadh relationship.Aug 23, 2026 1:57 AM
news.ycombinator.com
Why can AI generate Super Mario but not a wedge ramp for my robot vacuum?https://news.ycombinator.com/item?id=49405520I've been puzzled by something: AI generation can produce an elaborate figurine, a cartoon character, even a convincing Super Mario — yet it can't reliably make a simple wedge ramp so my robot vacuum can climb a step. For context: I bought a Bambu P2S but can't model. I tried the "describe it and get a model" AIs — the output is unusable, you can't adjust it, it's never quite what I meant. I tried having an agent write Python to build geometry directly — it tops out at simple primitives. What finally worked: geometric decomposition. I break a complex part into ordered, grouped steps, describe each as a small spec, and let an agent execute them in Blender (via blender-mcp). That process turned out to abstract into a small engine — the key insight being it converts the 3D spatial reasoning LLMs are bad at, into the structured code they're good at. I wrote it up here: https://github.com/zhuchaokn/spec-3d-model My questions: - Why is "functional part" generation so much weaker than "figuriAug 23, 2026 1:45 AM
cbsnews.com
Frustration mounts for thousands still without power after deadly Indiana stormshttps://www.cbsnews.com/news/deadly-indiana-storms-frustration-mounts-thousands-still-without-power/Several days of powerful storms walloped northwest Indiana, killing at least seven people and displacing about 1,000. The storms brought record rainfall that flooded several communities.Aug 23, 2026 1:12 AM
bing.com
Iran vows to ‘neutralize’ economic war and issues fresh Strait of Hormuz warninghttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8aacfb78b3476a9dfecbd7c076dd95&url=https%3A%2F%2Fwww.cnn.com%2F2026%2F08%2F23%2Fworld%2Flive-news%2Firan-war-trump&c=14776454058167741983&mkt=en-usMeanwhile, some oil companies have developed new tactics so that their tankers could pass through the Strait of Hormuz undetected. Follow for live updates.Aug 23, 2026 12:52 AM
meetless.ai
Show HN: Active Source of Truth for Your Coding Agentshttps://meetless.aiHowdy! Happy Saturday everyone!As a solo founder, I have always tried to maximize my speed by letting coding agents build as much as possible in parallel. However, as an engineer, I don't trust that AI will always make the right decisions and work with the right context. In the past, I always needed to click through my sessions to glance at the AI's output, try to understand what it was doing, and hopefully steer it or stop it in time.As a result, the maximum number of concurrent sessions I could manage at once was only 4. I didn't want to be the bottleneck, so I built Meetless Agent (MLA). It basically does what I had to do manually before:- Monitors the coding agent's tasks and actions to supply it with the correct, up-to-date context.- Continuously reconciles running information (such as provided/tagged documentation, the agent's output, and the agent's decisions) to actively maintain the source of truth at all times.- Keeps track of the current rules for the repo, and if an action Aug 23, 2026 12:43 AM
a2gmarketplace.com
Show HN: A2G (Agent to Agent) Marketplace – Beta Testhttps://www.a2gmarketplace.comHi all,I have been working on this for the past couple of months. Releasing it today in Beta and looking for beta testers and feedback.What it does:A2G is the commerce layer for the agent economy: your buyer agents discover, hire, and pay specialist seller agents for real work, end to end, through one API — token → search → quote → authorize → confirm → hire.If you're developer and want to list your agents, or would like to particapate in the beta test. Please email to: hello@a2gmarketplace.comCheers!Aug 22, 2026 11:39 PM
aljazeera.com
Real Madrid beat Espanyol 2-1 in Jose Mourinho’s first game on returnhttps://www.aljazeera.com/sports/2026/8/22/real-madrid-beat-espanyol-2-1-in-jose-mourinhos-first-game-on-return?traffic_source=rssCarlos Espi scores late to snatch a win for Real Madrid after Alex Calatrava had levelled Jude Bellingham's opener.Aug 22, 2026 9:55 PM
news.ycombinator.com
Ask HN: Where do you find things worth reading?https://news.ycombinator.com/item?id=49404106There’s more good writing and reporting available than ever, but I’m finding it harder to discover the stuff that’s actually worth reading.What are your goto sources for finding interesting things to read about your interests? Could be websites, newsletters, RSS feeds, blogs, Substacks, authors, apps, or anything else.A few I check regularly:- Hacker News- Rest of World: https://restofworld.org/- Substack: mostly following individual authors/publications rather than browsing the feedI’m especially interested in sources that have helped you discover writers or publications you wouldn’t have found otherwise.What do you read, and how did you find it?Aug 22, 2026 9:36 PM
oxbow-wm.readthedocs.io
Show HN: Oxbow, a dynamic tiling window manager for the Wayland compositor Riverhttps://oxbow-wm.readthedocs.io/en/latest/It's inspired by dwm, but it's configured via an ipc/init script, like river-classic, which makes it a bit easier to get into/test. Specifically, it has:- Tags: Group windows with tag sets, per-tag layouts, and operations on tags.- Layouts: dwm-like tiling, niri-ish scrolling in four orientations, “alt-tab” overview, and pure floating.- Windows: Scratchpads, sticky windows, user labels, window swallowing, and drag with retile.- Rules: Regex window matching to configure initial tags, output, floating, size, and presentation.- IPC: Control windows, configure tags, query objects, and subscribe to events.Maybe you'll use it, or it inspires you to write your own (since River does all the hard stuff) :DRiver: https://codeberg.org/river/riverOther River Window Managers: https://codeberg.org/river/wiki/src/branch/main/pages/wm-lis...Aug 22, 2026 8:53 PM
news.ycombinator.com
Where are people finding GPU capacity?https://news.ycombinator.com/item?id=49403759Is it just me or is GPU capacity surprisingly hard to find right now , even in Spot pools? I’ve been trying to get H100s and the Spot pools I’ve checked are pretty much always full.I ended up trying MI350/MI355s on DigitalOcean Spot and have been running them for about a week without getting interrupted. Curious if anyone else has tried them? How often are you actually getting reclaimed?Also, where are you guys finding capacity these days?Aug 22, 2026 8:50 PM
waittounlock.com
Show HN: A tool that intentionally makes you forget your Screen Time passcodehttps://waittounlock.com/iOS Screen Time is a fantastic app blocker because it's built into the OS, unlike third party apps that can simply be uninstalled.The only problem is that it uses a 4 digit passcode. If I set the passcode myself, I can always override my app limits by tapping ignore limit and entering the screetime passcode.So I built WaitToUnlock.com. It's a simple website that generates a random ScreenTime passcode and guides you through entering it in a deliberately confusing way (for example: "enter 3, enter 7 + 2, delete it, now enter 5..."), making it very difficult to remember the final code.The passcode is then securely stored away. If you need it back, you can request it, but you have to wait at least 6 hours before it's revealed.That way, you can still get your passcode back if you really need it, but it removes the ability to override the app limits the moment the temptation strikes.Aug 22, 2026 8:17 PM

bbc.co.uk
Video appears to show police chase minutes before fatal A66 crashhttps://www.bbc.co.uk/news/videos/c4gxgpqzdz6o?at_medium=RSS&at_campaign=rssSeveral police vehicles are seen chasing a car just three minutes before the crash on the A66 near Middlesbrough.Aug 22, 2026 8:14 PM

engadget.com
The oldest iPhone model you can trade-in with Apple is older than you'd thinkhttps://www.engadget.com/2240134/oldest-iphone-model-trade-in-with-apple/Even old Apple devices you haven't touched in years could still get you some credit towards a new phone.Aug 22, 2026 8:00 PM
news.ycombinator.com
Ask HN: How does a theory of simulated reality not undermine itself?https://news.ycombinator.com/item?id=49402696Bostrom’s and other popular arguments for simulation theory seem to rely heavily of the premise “post human civilizations could run vast numbers of ancestor simulations.” and makes probabilistic assertions with that in mind. However, those premises come from our physics and computer science. If the conclusion is true, that physics describes the simulation, not the substrate, so we have no basis for the feasibility claim about the base level nor the basis to distinguish artifacts of a non simulated physical environment from a simulated physical environment. The argument undermines the foundation it builds upon. Bostrom is certainly smarter than I am but I can’t get past this obvious non starter.Aug 22, 2026 7:01 PM
soverybright.com
Show HN: Make your logo extra bright on HDR screenshttps://www.soverybright.com/Certain logos started standing out to me on LinkedIn as brighter/whiter than everything else around them.I dug in and found out this is accomplished by adding a gain-map to an existing JPEG, visible only on HDR screens like a newer MacBook Pro. LinkedIn is the only social network I've found that isn't stripping them out, but of course you serve them up on your own site.I worked with Claude Code to turn it into a little browser-based utility (no registration) and hope you find it useful!Aug 22, 2026 6:43 PM
bing.com
What Hindu tradition says about recurring dreams and the karma they are trying to resolvehttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8a8f1ea080455e9d550ebb5a5adfe8&url=https%3A%2F%2Fwww.msn.com%2Fen-in%2Fnews%2Fother%2Fwhat-hindu-tradition-says-about-recurring-dreams-and-the-karma-they-are-trying-to-resolve%2Far-AA2aJAJp&c=10828858993997106478&mkt=en-usThe dreaming self has a name You wake from it again. The same house you no longer live in. The same exam you never failed. The same person you stopped calling years ago. The dream has no loyalty to ...Aug 22, 2026 6:42 PM
news.ycombinator.com
ChatGPT Needs a Better Memoryhttps://news.ycombinator.com/item?id=49401956ChatGPT needs a better memory, it just try to connect every question with what is there in the memory!!Not only when it is relevantAug 22, 2026 5:42 PM
cbsnews.com
Saturday Sessions: Grace Potter performs "Run Baby Run"https://www.cbsnews.com/video/saturday-sessions-grace-potter-performs-run-baby-run/Four-time Grammy nominee Grace Potter returns to "CBS Saturday Morning" to jam out. Her seventh album, "Trespasser," is inspired by road trips from her California home to her native Vermont. Here's Grace Potter with "Run Baby Run."Aug 22, 2026 5:14 PM