Forem: David Cassel

Sam Altman wonders: Could the government nationalize artificial general intelligence?

David Cassel — Thu, 05 Mar 2026 19:01:28 +0000

He said it! "It has seemed to me for a long time it might be better if building AGI were a government project," Sam Altman said this week. While he may not think it's likely, “I do think a close partnership between governments and the companies building this technology is super important.”

Trump's War Department even threatened a kind of "soft nationalization" against Anthropic, according to Fortune magazine's AI editor, by suggesting they'd invoke 1950's "Defense Protection Act" to force Anthropic to accept the government's terms for military contracts.

Altman weighs in - along with OpenAI's Head of National Security Partnerships - actually fielding live questions from Twitter users at one point. It proves the community of AI users has a role to play - by their software choices, but also by engaging in all the thoughtful discussions that are happening.

When a 20-Year-Old Bill Gates Fought Free Software 'Piracy'

David Cassel — Sun, 01 Feb 2026 17:42:23 +0000

As the first home computers appeared in the mid-1970s, the world faced a question: Would its software be free? Bill Gates -- then 20 years old -- screamed no.

“Most of you steal your software…” wrote 20-year-old Gates. “Who cares if the people who worked on it get paid?” Gates had coded the BASIC interpreter for Altair's first home computer (with Paul Allen and Monte Davidoff) -- only to see it pirated by Steve Wozniak's friends at the Homebrew Computing Club.

Expecting royalties, a none-too-happy Gates issued an "Open Letter to Hobbyists" in the club's newsletter (as well as Altair's own publication), complaining "Who can afford to do professional work for nothing...? I would appreciate letters from any one who wants to pay up." But freedom-loving coders had other ideas.

When Steve Wozniak and Steve Jobs released their Apple 1 home computer that summer, they stressed that "our philosophy is to provide software for our machines free or at minimal cost..." And the earliest open-source hackers began writing their own free Tiny Basic interpreters to create a free alternative to the Gates/Micro-Soft code. (This led to the first occurrence of the phrase “Copyleft” in October of 1976.

And ultimately Gates/Micro-Soft gave up and abandoned royalty-based pricing for future software they wrote for Altair, switching to up-front payments from the company.

Linus Torvalds Gets Candid About Windows, Workflows, and AI

David Cassel — Sun, 04 Jan 2026 15:55:32 +0000

Linus Torvalds recently surprised his fans with a surprisingly fresh perspective on how Linux gets built, appearing on the YouTube channel Linus Tech Tips to build a custom PC - and shared some genuinely new insights into the mind of the man who's led the Linux project since 1991.

With millions relying on his work, Torvalds said he can't have his hardware making him think the OS has bugs. "I'm convinced that all the jokes about how unstable Windows is and blue screening… a big percentage of those were not actually software bugs. A big percentage of those are hardware being not reliable."

And along the way, Torvalds improvised some clever answers — like when asked how he handles data storage. "I upload it on the internet, and if it's worth saving, somebody else will save it for me!"

He also revealed his favorite videogame, what he thinks of the GPL license, and even a relationship with Microsoft that he know thinks of as "friendly"!

Adafruit: Arduino's Rules are 'Incompatible with Open Source'

David Cassel — Sun, 14 Dec 2025 18:36:51 +0000

"Last week, I spoke to Arduino, Adafruit and the EFF about Qualcomm's October acquisition of the beloved company known for its single-board microcontroller kits" -- and, also, its new Terms & Conditions.

Adafruit founder Ladyada says Arduino’s usage restrictions “effectively override the freedoms the license is supposed to guarantee.” Arduino now lists “prohibited uses of AI," which include criminal use and violation of the law, intentions to harm (including dissemination of false information and manipulative or deceptive acts), generating facial recognition databases and military use.

An Arduino blog post argues its new AI features are optional. But their conditions also include the right to "monitor User accounts and use of the AI Product … [for] verifying compliance with laws and this policy.” (And for other reasons, including “administering and managing Arduino’s business.”)

Adafruit’s Fried says Arduino “should, of course, comply with applicable laws and respond appropriately to clear evidence of criminal activity.” But “they should design their AI and cloud offerings so that monitoring is narrowly targeted, proportionate, and clearly explained, instead of defaulting to broad surveillance across all users.”

And that's also where it stops being open source. “Genuine open source licenses do not allow field-of-use restrictions,” Fried said. “You cannot say ‘this code is open source, but it may not be used for military purposes’ and still call the license open source."

2025's 'Advent of Code; Event Chooses Tradition Over AI

David Cassel — Sun, 30 Nov 2025 16:44:32 +0000

AI may be changing the world of programming, but some human holiday traditions continue. Hundreds of thousands of coders await this year's "Advent of Code" event with its daily holiday-themed programming puzzles.

And this year event creator Eric Wastl has made some more changes to discourage using AI. The site's FAQ list now specifically tells users that they shouldn't use AI when solving puzzles. “If you send a friend to the gym on your behalf, would you expect to get stronger?" And he's also removed the global leaderboard - a change long overdue, according to one a response he received from a commenter on social media. "I thought the global leaderboard should've gone a few years ago anyways, when LLMs started being a thing."

Microsoft EVP: Embrace AI Agents to Rewire Business Processes Now

David Cassel — Wed, 19 Nov 2025 20:26:17 +0000

The executive VP overseeing Microsoft's AI-powered assistant Copilot runs a $100-billion business with 30,000 employees. And he warns leaders that AI agents are "not some distant future. It’s happening now."

He wants everyone to know that “It’s possible today to take full advantage of agents.” Appearing on a Microsoft podcast, he stressed that “The security model exists, the identity model exists, the user interface exists. The hard work here is to actually go pick the processes that give you the most bang for the buck — and then be rigorous about measuring that.”

And that's why there's the Copilot impact dashboard — "so customers can take their core key performance indicators, and they can measure how Copilot is moving those key performance indicators."

A Go Expert Complains: 'I Don't Want to Maintain AI-Generated Code'

David Cassel — Sun, 28 Sep 2025 17:11:34 +0000

It was a moment for our time — two long-time Go programmers pondering the future of their language, and how it will be impacted by AI...

"The bar to accept deliverables is very low..." said Quebec-based programmer Dominic St-Pierre, who's been building software systems for the last two decades. And after running some tests with Claude, "I'm sorry, but it's not good code. And I would not want to be in a team that produced that kind of code." Laughing, he said, "I don't want to maintain code that the AI generates, because it's mostly not good."

During an interview on St. Pierre's podcast, Go author John Arundel pushed back. Is bad software produced by AI any worse than bad software produced by humans? St. Pierre believes "The amount of things that the AI is outputting is overwhelming!"

But Arundel says "I've decided that I'm going to feel optimistic, reasoning this is going to bring lots of people into the world of programming, which is a fun world..."

Java Language Architect Brian Goetz on How Java Could Evolve

David Cassel — Sun, 21 Sep 2025 17:31:47 +0000

Java language architect Brian Goetz spoke at last month's JVM Summit, delivering a talk that looked to Java's future.

Goetz discussed not the Java we have now, but a hypothetical "set of features that are designed not to be used by themselves as a way to write better programs — but as a mechanism for making the language more growable and more extensible."

In short, Goetz explained how he sees the Java language evolving, "and we feel like we've come to a point now where we have a pretty good idea of which way we want to go with this."

OpenAI's Sam Altman Sees a Future Beyond GPT-5 With Collective 'Superintelligence'

David Cassel — Thu, 11 Sep 2025 22:53:30 +0000

Just one week after GPT-5 launched, OpenAI's Sam Altman shared the grandest thought of all: that maybe it's society that's the superintelligence. "No one person could do, on their own, what they're able to do with all the really hard work that society has done together to give you this amazing set of tools." In a mid-August interview on YouTube, Altman argued that it's the kids born today who will ultimately inherit what a superintelligent society has created for them.

So what is the sum of all technologies? If our future harnesses everything ever built, by everyone who came before us — how does it culminate?

Sam Altman has some thoughts...

What Unix Co-Creator Brian Kernighan Thinks of Rust

David Cassel — Sun, 31 Aug 2025 20:24:30 +0000

83-year-old Brian Kernighan co-created Unix - and recently amused an audience with the troubles he'd had trying Rust. "I found it a pain!" Kernighan said at the "Vintage Computer East" festival.

"I just couldn’t grok the mechanisms that were required to do memory safety, in a program where memory wasn’t even an issue!"

But it wasn't just that. "The support mechanism that went with it — this notion of crates and barrels and things like that — was just incomprehensibly big and slow! And the compiler was slow, the code that came out was slow..."

"I have written only one Rust program, so you should take all of this with a giant grain of salt," Kernighan cautioned his audience. (Acknowledging later that "I'm probably unduly cynical.")

"But I'm — I don't think it's gonna replace C right away, anyway."

Guido van Rossum Revisits Python's Life in a New Documentary

David Cassel — Thu, 28 Aug 2025 18:40:28 +0000

A new documentary about Python interviews Guido van Rossum about the early days, his resignation as "Benevolent Dictator" - and everything in between. (Including his love for Monty Python...)

"It was a massive undertaking," the director explains in a new interview - adding "It feels great to finally release it to the world!"

David Heinemeier Hansson on Vibe Coding, AI, and Programming's Future

David Cassel — Sun, 27 Jul 2025 22:03:49 +0000

He's passionate about programming and productivity. So how does David Heinemeier Hansson (DHH) feel about AI-assisted vibe coding?

For his 474th podcast, Lex Fridman performed a special six-hour interview with DHH, the legendary creator of Ruby on Rails, who proved to be a shrewd observer of our moment in time. In short, DHH says he loves collaborating with AI: for creating drafts, looking up APIs or even getting a second opinion. But he uses it differently than most people, always keeping his AI-generated code in a separate window. Otherwise "I can literally feel competence draining out of my fingers!"

It's possible that one day, AI-assisted programming will mean more programming gets done, paradoxically increasing the amount spent on human programmers. But DHH says he's also open to the possibility that AI could do to programming what cars did to horses — turn programming into something we only turn to recreationally