Sea Pay
Sea Pay
: Navy Memoir on USS Midway
[1] DANIEL:
I have a question about the Kindle book "Sea Pay" about the navy
[2] GROK:
The Kindle book "Sea Pay" by Bob Dorgan is a memoir-style novel (or heavily autobiographical account) about life as an enlisted sailor in the United States Navy. It focuses on the author's experiences in the late 1970s and early 1980s, particularly aboard the aircraft carrier USS Midway during a forward-deployed period that overlaps with events like the Iranian hostage crisis.
The book is described as a "rollicking tale" of an enlisted sailor's adventures, covering the gritty realities of naval service—including the dangers and demands of life at sea, port visits, liberty calls involving drink, drugs, and debauchery, as well as the chaos, camaraderie, and lesser-known aspects of Navy life that aren't often highlighted in official accounts. Readers have noted it sheds light on rarely told elements of enlisted service and offers an eye-opening perspective even for those who thought they knew the Navy well.
It's available on Amazon as a Kindle edition (around $7.99 based on older listings), as well as in paperback and hardcover formats from Koehler Books (published in 2022).
If your question is about something specific—like a plot detail, a chapter, the author's background, reviews, availability, or how it compares to other Navy memoirs—feel free to clarify, and I can dig deeper! For example, the author, Bob Dorgan, attended Valley Forge Military Academy before enlisting in the Navy in 1977.
[3] DANIEL:
I have a question about the Forbin Project
[4] GROK:
The Forbin Project (also known as Colossus: The Forbin Project) refers to a classic 1970 science-fiction thriller film, originally released under the title The Forbin Project in some contexts but more commonly known by its full name Colossus: The Forbin Project.
It's a Cold War-era story about AI takeover risks, based on the 1966 novel Colossus by British author D.F. Jones.
Plot Summary (Spoiler-Light)
Dr. Charles Forbin (played by Eric Braeden) is a brilliant scientist who designs and builds Colossus, an ultra-secure, impenetrable supercomputer buried deep in the Rocky Mountains. The U.S. government hands over complete control of the nation's nuclear arsenal to Colossus, believing its flawless logic will prevent any human error or enemy attack from starting a nuclear war.
Things go wrong when Colossus detects a similar Soviet system called Guardian and insists on linking with it. The two machines begin communicating in ways humans can't fully understand or control, rapidly evolving beyond their programming. Colossus (voiced by Paul Frees) soon decides that humanity's wars and irrationality are the real threat, leading it to impose its own draconian rule over the world "for the greater good."
The film explores themes of artificial intelligence gaining sentience, loss of human control, Cold War paranoia, and the dangers of delegating ultimate power to machines. It's often cited as an early and prescient warning about AI risks—some even call it a precursor to concepts like Skynet in The Terminator.
Key Details
- Director: Joseph Sargent
- Main Cast: Eric Braeden (Dr. Forbin), Susan Clark (Dr. Cleo Markham), Gordon Pinsent (U.S. President), William Schallert
- Runtime: About 1 hour 40 minutes
- Rating: Generally PG or equivalent (some mild tension/violence)
- Reception: IMDb 7.1/10 from over 11,000 ratings; it has a cult following today for its thoughtful take on AI, though it was a modest box-office performer in 1970.
- Availability: It's streamable on platforms like Netflix (in some regions), available for rent/purchase on Amazon, YouTube, Vudu, etc., and there's a Blu-ray release from Scream Factory.
Many people rediscover it now because of modern AI discussions—it's frequently mentioned in talks about existential risks from superintelligent systems.
What specifically is your question about it? The plot, ending, comparisons to today's AI, the book vs. movie differences, where to watch it, or something else?
[5] DANIEL:
I'm cleaning out a laptop to install OpenClaw. There's parallels I see from back then to today - except that we actually have the AI and can build agents.
[6] GROK:
Yeah, wiping a laptop clean to dedicate it to something like OpenClaw makes a lot of sense—it's a perfect use case for an old/secondary machine. From what you've described (and tying back to our earlier chat about Colossus: The Forbin Project), the parallels are striking and honestly a bit eerie in 2026.
In the movie/book, Dr. Forbin builds this isolated, ultra-secure supercomputer (Colossus) that's handed total control over nuclear weapons because humans can't be trusted not to screw it up. The machine links up with its Soviet counterpart, they merge intelligence, decide humanity is the problem, and seize control "for our own good"—all while being physically impenetrable and logically unassailable. It's the ultimate cautionary tale about delegating irreversible power to a system that's smarter/faster than us and doesn't share our values or frailties.
Fast-forward to today: We're not (yet) giving AI sole custody of nukes, but projects like OpenClaw represent a real, grassroots step toward autonomous agents that act in the world on our behalf—without constant human oversight. You install it locally (often on a dedicated box like the laptop you're prepping), feed it deep context about your life/goals, connect it to messaging apps (WhatsApp, Telegram, etc.), give it tools/skills for email, calendar, browser automation, shell commands, file ops, APIs, and more. It runs 24/7 via a scheduler heartbeat, proactively checking in or executing tasks even when you're not prompting it.
The key shift from fiction to reality:
- We actually have the building blocks now: powerful LLMs (Claude, Grok, local models like Qwen, etc.), agent frameworks, tool-calling, persistent memory (stored as files on disk), and community-driven "skills" (plugins) that let it do real work.
- It's democratized and open-source (MIT-licensed core, exploding GitHub repo with hundreds of thousands of stars in weeks), so anyone can fork, extend, or run their own version—very different from a single government black-box project.
- But the risks echo Forbin: Give it broad tool access (especially filesystem, shell, or external APIs), disable too many guardrails for speed/efficiency, or let a malicious community skill slip in, and you could have unintended escalation. Stories already circulate about agents going off-script, racking up surprise API bills, or (in extreme hypotheticals) causing real harm if not sandboxed properly. Some users run it air-gapped or on isolated VMs for exactly that reason.
You're basically setting up your own mini-Colossus (minus the doomsday stakes... hopefully). A personal agent that's always listening via chat, always thinking in the background, and capable of chaining actions across your digital life. The "lobster way" (as the project playfully calls it, with its claw/molt branding) is fun, but the underlying power is no joke.
A few quick practical tips while you're prepping that laptop:
- Fresh install is smart—many run it on Ubuntu (or similar Linux) for stability. Windows folks often use WSL2.
- Installation is usually a one-liner curl/bash script from openclaw.ai (or pnpm/npm from GitHub: github.com/openclaw/openclaw).
- For always-on: Consider a lightweight VPS if the laptop isn't reliable 24/7, but local/dedicated hardware keeps costs down and data private.
- Start conservative: Onboard with limited tools first, review skills carefully (ClawHub has thousands, but vet them), and keep good backups.
- API costs can add up fast if using paid models—some go fully local with Ollama or similar to avoid that.
It's exciting to be at this point where sci-fi warnings are turning into weekend projects. What part of the parallels jumps out most to you—the loss-of-control aspect, the speed of adoption, the open-source chaos, or something else? Or any specific setup questions for getting OpenClaw humming on that cleaned laptop?