EV models 2006 - 2027

 Every NEW EV Coming In 2026 - 27 | Tesla, Rivian, VW, BMW, Honda, & More - YouTube @ Ryan Shaw - YouTube

The EV market is undergoing a major shift. Here is a breakdown of the models mentioned in the video that have been discontinued or canceled, followed by the new vehicles launching in 2026 and 2027.

Discontinued and Canceled EVs

• Honda: Zero Series (Saloon/SUV), Acura ZDX, Honda Prologue (1:05-1:46)
• Apple: Project Titan (1:47-2:01)
• Fisker: Ocean SUV (2:02-2:15)
• Tesla: Model S and Model X (2:16-2:28)
• Ford: F-150 Lightning (pure electric version) (2:29-2:51)
• Volvo: EX30 (US market) (2:52-3:08)
• Mercedes-Benz: EQE (sedan/SUV), EQS (sedan/SUV) (3:09-3:26)
• Chrysler: Airflow (3:27-3:30)
• Ram: 1500 REV (long-range version) (3:31-3:42)
• Volkswagen: ID7, ID Buzz, ID4 (production winding down) (3:43-3:53)
• Nissan: Ariya (paused US sales) (3:54-3:56)
• Genesis: Electrified G80 (3:57-3:59)
• Kia: Niro EV (dropping), EV6/EV9 GT (performance trims delayed) (4:00-4:07)

New and Upcoming EVs (2026–2027)

• Tesla: Model Y (2026 refresh), Model YL (5:56-8:18)
• Rivian: R2, R3, R3X (8:39-12:00, 19:47-20:53)
• BMW: iX3 (12:01-13:46)
• Porsche: Cayenne Electric (13:47-15:16)
• Chevrolet: Equinox EV, Bolt (new generation) (15:17-16:26, 30:16-30:45)
• Kia: EV3 (16:27-17:53)
• Nissan: Leaf (2026 redesign) (17:54-18:37)
• Mercedes-Benz: C-Class Electric, CLA Electric (18:38-19:46, 22:29-23:50)
• Lucid: Cosmos, Earth, Gravity (20:54-22:28, 27:26-28:29)
• Hyundai: Ioniq 9, Ioniq 5N (23:51-24:59, 33:22-33:45)
• Audi: Q6 e-tron (25:00-26:17)
• Polestar: Polestar 3, Polestar 4 (26:18-27:25)
• Volvo: EX90, EX60 (28:32-29:41)
• Subaru: Trailseeker (29:42-30:15)
• Scout Motors: Terra, Traveler (30:46-31:21)
• Toyota: Highlander EV, Corolla (potential EV) (31:22-32:02, 33:02-33:21)
• Lexus: TZ (32:03-32:13)
• Ford: Midsize electric pickup (32:14-32:41)
• Volkswagen: ID Tiguan (32:42-33:01)

Color Circle

Colour - Visible Spectrum, Wavelengths, Hues | Britannica

One form of Newton's colour circle.

Color | Definition, Perception, Types, & Facts | Britannica

Fisker => open source car company

 Fisker went bankrupt and owners built open source car company from the ashes | Electrek

When Fisker Inc. filed for Chapter 11 bankruptcy in June 2024, it left roughly 11,000 Ocean SUV owners holding the keys to vehicles that cost them anywhere from $40,000 to $70,000 — and that were rapidly losing the software brains that made them work. No more over-the-air updates. No more connected services. No more warranty. The manufacturer was dead.

What happened next is one of the most remarkable stories in the history of the electric vehicle industry. Instead of accepting that their cars would become rolling paperweights, Fisker Ocean owners organized, reverse-engineered their vehicles’ proprietary software, hacked into CAN bus networks, built open-source tools on GitHub, and effectively stood up a volunteer-run open-sourced car company from the ashes of Fisker.

Within months of the bankruptcy filing, thousands of Ocean owners formed the Fisker Owners Association (FOA) — a nonprofit that quickly grew to 4,000 members and began operating as something between a car club, a tech startup, and an independent automaker.

 

AI: export Google AI Studio chats

 in google ai studio, is there a way to copy/export whole conversation?

---

Currently, Google AI Studio does not have a native "Export" or "Download" button for saving your conversations in readable formats like PDF or Markdown.

Furthermore, simply hitting Ctrl+A or Cmd+A to copy the page manually usually fails on longer chats. This is because AI Studio uses "virtual scrolling," meaning the interface only loads the chat bubbles currently visible on your screen.

However, there are three practical ways to export or copy your entire conversation, depending on how you need the data:

1. The Best Method: Use a Browser Extension (Third-Party)

Because this is a major pain point for developers, the community has created free Chrome extensions specifically designed to bypass the virtual scrolling limitation and extract the full chat.

• Google AI Studio Exporter or Chat Exporter for Google AI Studio: These are free extensions in the Chrome Web Store. When clicked, they automatically scroll through your entire chat to capture every message and let you download the conversation as a clean Markdown (.md), PDF, or plain text file.
• Gemini to PDF: Another popular community extension that adds a "Save Chat" button to export your conversations preserving formatting, code blocks, and tables.

2. The Quick Native Workaround: The "Get Code" Button

If you don't want to install an extension and just need the raw text of the entire conversation, you can use the built-in code generator:

1. Click the Get Code button at the top right of the AI Studio interface.
2. Switch the tab to cURL, JSON, or Python.
3. Inside the code snippet, look for the contents or history array. This contains your entire back-and-forth conversation in a structured text/JSON format.
4. You can copy this entire block of code to your clipboard. While it includes coding syntax, it ensures no part of the conversation is cut off.

3. The Backup Method: Download the JSON from Google Drive

If you have "Auto-save" enabled, Google AI Studio automatically backs up your chats to your Google Drive.

1. Go to your Google Drive and locate the saved AI Studio file (it usually has the blue/white AI Studio butterfly icon).
2. If you double-click it, it will just reopen AI Studio.
3. Instead, right-click the file and select Download.
4. This will download a .json file containing the complete chat history, prompt configurations, and system instructions, which you can open in any text editor or parse with a script.

AI "SaaSpocalypse"

SaaSpocalypse - Wiktionary, the free dictionary

A predicted (and disputed) implosion of established SaaS development companies as a result of emerging commercialized AI being fundamentally disruptive to the industry.

Some investors are quick to dump SaaS stocks because they predict a SaaSpocalypse, whereas others believe that AI is changing how SaaS companies do their work without making those companies unnecessary or unviable.

Related terms AIpocalypse

 Trump-Xi Summit, Benioff: "Not My First SaaSpocalypse," OpenAI vs Apple, Multi-Sensory AI, El Niño - YouTube

The conversation highlights several critical insights regarding the current state and future of artificial intelligence within the software and technology sectors:

• The 'SaaSapocalypse' and Rerating: The enterprise software market is undergoing a significant correction, often described as a 'SaaSapocalypse.' While top software companies are still posting strong quarterly results, their valuations have been rerated downward as markets adjust their expectations in light of the AI revolution.
• The Shift to Agents and Platforms: There is a fundamental shift occurring from traditional, static software to more dynamic systems. Businesses are moving toward headless platforms, where AI agents interoperate to automate complex tasks, such as outbound sales and customer service, far more efficiently than previously possible.
• The Importance of Context and Data: AI models are probabilistic and require high-quality, 'grounded' data to be effective in an enterprise setting. Integrating data across platforms (e.g., Salesforce, Slack) is essential for AI to act as a reliable 'world model' for businesses.
• The Evolution of Coding: Coding has become vastly more efficient due to generative AI, enabling non-technical users to build and interact with complex systems. However, this has also led to intense competition for talent and a race among AI labs to pivot toward the most effective tools, with a current focus on coding agents.
• Token Efficiency: While there is fear that the adoption of real-time, multi-sensory AI models will lead to unsustainable costs, there is a counter-argument that current token usage is inefficient. The industry is expected to develop intermediary layers to route tasks to the most affordable model available rather than defaulting to the most expensive ones.
• Hardware and Local Models: There is a strong trend toward localized AI running on high-end hardware, such as new Mac processors. This approach provides a clear path to maintaining privacy and reducing reliance on cloud-based AI, which faces growing trust issues.

stock price charts can be misleading, depending how one looks at them... context matters

Architecture, AC without HVAC? "cross ventilation"

This FREE Hack Will INSTANTLY Cool Your Home By 42F — Here’s How. - YouTube

This video explores the science and history of cross ventilation, a design principle used for centuries by civilizations in the Middle East, Africa, and Asia to create survivable indoor climates in extreme heat. We dive into the physics of the Bernoulli principle, explain how architectural icons like the Iranian Badgir and the Southern Shotgun house acted as natural wind tunnels, and reveal why modern building codes have moved away from these efficient designs. 

Whether you're looking to install a whole-house fan or simply want to know how to reposition your furniture and windows to drop your indoor temperature by up to 6∘ F, this guide provides the tools to outsmart the summer heat without touching your thermostat.

AI: Dario Amodei & Marc Benioff, Anthropic & Salesforce

very informative and uplifting chat

Dario Amodei & Marc Benioff: Future of AI - YouTube @ Salesforce Events - YouTube

Dario Amodei | LinkedIn

In this discussion, Dario Amodei co-founder and CEO of Anthropic,

and Marc Benioff, founder and CEO of Salesforce, and early investor in Anthropic,

cover the evolution of Anthropic, the future of AI, and its integration into the enterprise.

Key Points:

• Founding Philosophy: Dario Amodei and his seven co-founders, including his sister Daniela Amodei, left OpenAI to establish Anthropic with a core focus on safety, responsibility, and the development of reliable models for enterprise use cases.
• The Evolution of Coding: A significant shift is occurring where AI models are now capable of writing large portions of code. Rather than replacing software engineers, this is creating a shift toward rebalancing roles where engineers act as supervisors and editors, significantly increasing their productivity.
• Agentic Capabilities: The industry is moving from simple text-generation toward "agentic" models that can perform end-to-end tasks, such as managing workflows or analyzing complex medical data, acting more like autonomous coworkers.
• Enterprise Transformation: Anthropic is positioning itself to serve the enterprise world, focusing on industries like finance, healthcare, and insurance, where trustworthiness and reliability are paramount.
• Infrastructure and Compute: The massive demand for compute power to support scaling laws remains a primary driver for data center expansion, though Amodei cautions that public discourse around data center deals may sometimes be inflated by "double counting."

Notable Quotes:

• "Building ordinary software, it's like building a skyscraper...
  This (AI) is ironically a bit more like biological. It's organic...
  You're growing these models." — Dario Amodei
• "What is an enterprise? It's kind of a super intelligence. It is an entity that acts in the world in a way with a level of strategy and knowledge and power that goes far beyond what any individual human can do." — Dario Amodei
• "I'm optimistic about complementarity in the short run. I do worry as we get out to 2 years, 5 years that across the economy... we'll have a kind of labor disruption that is not fundamentally different from what we've seen with previous technologies but that operates faster." — Dario Amodei

Marc Benioff Said Salesforce Will Spend $300M on Anthropic Tokens - Business Insider

The Making Of Dario Amodei - by Alex Kantrowitz

The profile explores the motivations, background, and business trajectory of Dario Amodei, the co-founder and CEO of Anthropic. It highlights how a personal tragedy—the loss of his father to a rare illness that became largely curable just a few years later—fuels his urgent drive to accelerate AI development to solve massive biological and human problems. Despite being labeled by some as an industry "doomer" due to his outspoken stance on safety and regulation, Amodei considers himself a capability bull who advocates for safety frameworks specifically so that development doesn't have to slow down.

---

Key Points

• The Catalyst: Amodei’s shift from theoretical physics to biology, and eventually to AI, was profoundly shaped by his father’s death in 2006. He believes faster technological and scientific progress could have saved his father, driving his current short timeline for AI advancement.

• Pioneering Scaling Laws: During his time at Baidu in 2014, Amodei contributed to the formulation of AI "scaling laws"—the principle that predictably increasing data, compute, and model size yields better performance without needing entirely new methodology. He remains one of the industry's purest believers in this exponential growth curve.

• The OpenAI Split: Amodei led the landmark GPT-3 project at OpenAI. However, systemic disagreements over safety, commercialization, and corporate governance led Amodei and a core group of researchers (referred to as "the pandas") to break away and found Anthropic in late 2020.

• Massive Revenue Growth: Anthropic has become one of the fastest-growing software companies in history. Its annualized recurring revenue rocketed from $1.4 billion in March 2025 to roughly $4.5 billion by July 2025, primarily driven by businesses utilizing its models via API integrations and specialized coding tools like Claude Code.

• Financial & Safety Pressures: Despite massive revenue, Anthropic is deeply unprofitable, with projected losses of $3 billion for the year due to immense computing and training costs. To keep scaling, the company has raised nearly $20 billion from tech giants like Amazon and Google, and is exploring a new $5 billion round that includes Middle Eastern Gulf state investors.

• "Race to the Top" Strategy: Amodei pushes for strict export controls, safety regulations, and public stress-testing frameworks (like Anthropic's Responsible Scaling Policy). Rather than trying to create regulatory capture to shut out open-source competitors like DeepSeek, Amodei asserts he is trying to force an industry-wide upward trajectory in safety practices as models approach human-level knowledge work.

js tool: FluidCAD: Parametric CAD

 Parametric CAD for everyone | FluidCAD

Write JavaScript. See 3D geometry in real time.

Fluid-CAD/FluidCAD: Parametric cad modeling with Javascript @GitHub

TS, LGPL

antfly: AI DB in Go

Building a Distributed Search Engine in Pure Go — Antfly Blog

Antfly is a distributed document database and search engine written entirely in Go. It combines full-text search (BM25), vector similarity, sparse vectors (SPLADE), and graph traversal in a single binary. It also ships with built-in ML inference for embeddings, reranking, and chunking — so you can go from raw documents to production RAG without reaching for external APIs.

Run antfly swarm and you get a fully functional node with search, ML inference, a dashboard, and an MCP server — all in one process. Need to scale? Add more nodes and Antfly handles shard splitting and replication automatically.


antflydb/antfly @GitHub

Antfly is a distributed search engine built on etcd's raft library. It combines full-text search (BM25), vector similarity, and graph traversal over multimodal data — text, images, audio, and video. Embeddings, chunking, and graph edges are generated automatically as you write data. Built-in RAG agents tie it all together with retrieval-augmented generation.

Go 1.26 introduced the experimental simd/archsimd package, giving Go programs native access to SIMD instructions. Antfly uses go-highway — a portable SIMD library inspired by Google's Highway — to write vector operations once and run them on AVX2, AVX-512, ARM NEON, or a pure Go fallback.

MDN: React => Markdown & Web Components

 Under the hood of MDN's new frontend

EV: Tesla Model S legacy

 Tesla Never Stopped Developing The Model S — Revelations with Jason Cammisa - YouTube

The Tesla Model S: A Paradigm Shift

• Industry Transformation: The Model S is widely regarded as the most significant car of the last 75 years, having catalyzed the global shift toward electrification and forced traditional automakers to rethink their approach to vehicle development.
• Constant Iteration: Unlike traditional manufacturers that freeze designs for years, Tesla continuously updated the Model S with thousands of hardware and software improvements throughout its 14-year lifespan. As a result, the final version shares only about 3% of its parts with the original 2012 model.
• Software-Defined Vehicle: The car pioneered over-the-air (OTA) updates, allowing Tesla to fix issues, improve efficiency, and add new features—such as Sentry Mode—long after the customer had taken delivery. This redefined the car as a living, evolving piece of technology rather than a static product.
• Efficiency and Performance: By focusing on "battery bucks"—optimizing every component to save weight and electricity—Tesla engineers dramatically increased efficiency and performance over time. The modern Plaid variant is significantly faster, more efficient, and better equipped than early models, all while remaining competitively priced when adjusted for inflation.
• Challenging Conventions: While early models faced criticism for build quality, the Model S proved that consumers valued technological innovation, usability, and performance over traditional manufacturing metrics like panel gaps. It set the standard for what a desirable EV could be, paving the way for the Model Y to become the world’s best-selling passenger car.
• A Legacy of Progress: The video argues that traditional automakers who abandoned or delayed their EV programs in response to short-term market fluctuations face an existential risk. Tesla’s success has established a new industry reality where rapid iteration, software integration, and a focus on long-term efficiency are essential for survival.

Go => C; => GoLang

 Solod

solod/doc/spec.md at main · solod-dev/solod

Solod (So) is a strict subset of Go that translates to regular C — with zero runtime, manual memory management, and source-level interop.

• Go in, C out. You write regular Go code and get readable C11 as output.
• Zero runtime. No garbage collection, no reference counting, no hidden allocations.
• Everything is stack-allocated by default. Heap is opt-in through the standard library.
• Native C interop. Call C from So and So from C — no CGO, no overhead.
• Go tooling works out of the box — syntax highlighting, LSP, linting

ubavic/compiles-to-go: List of programing languages that compile to Go @GitHub

Go’s simplicity, fast compilation, and mature ecosystem make it an appealing compilation target for new languages. This is a list of languages that, in one way or another, compile down to Go assembly.

Golang Weekly Issue 596: April 10, 2026

A Fast Immutable Map in Go – Daniel Lemire's blog

x1unix/go-playground: Better Go Playground with autocompletion and syntax highlight

Better Go Playground