Adrian Mouat

Apropos of nothing, reply to this with your favorite robotic slur

I read somewhere once that the original pronunciation of ChatGPT was supposed to be chat gippity.

So now my favorite robot slur is instead of ChatGPT, sometimes I’ll say chat gippity do dah

I think you can do that?

I'm super excited for Wassette, so I put together a quick article on why it's important and a comparison with our current options for MCP.

Nice!

Not sure if I'm misunderstanding, but you can mount a single file as a volume in Docker.

This looks cool -- fan sites with chat powered by bluesky (or bluesky tech at least).

Yeah, pretty much. I am using readwise reader but it's not worth the money for all I use it.

I also found readeck, which is probably a much better solution, but I wanted to see how far I could go with this.

It just sends URLs (or data) to a service which runs them through readability and saves them to disk, so I can read them later.

Last step is figuring out some caching functionality for reading offline.

Which also makes me wonder if AI is any good at updating unmaintained software....

AI is going to lead to an explosion of personalised, unique tools like this.

It's awesome and I'm happy to see what happens, but there's going to be an awful lot of abandoned, unmaintainable software as well.

Just had Claude Code write a Chrome Plug-in for a stupid little app of mine.

It would have taken me ages to figure out how Chrome plug-ins work -- long enough that it wouldn't have been worth the effort for this use case. But with Claude Code it was done and working in seconds.

But the rest of the video is about how to go beyond this by creating MCP servers that match user intent.

The title is a bit clickbaity - it opens by pointing out there's no point in creating MCP integrations that are 1-1 copies of CLIs or APIs; if there's adequate documentation the LLMs will do a better job of calling the tool directly than using MCP.

YouTube video by AI & DevOps Toolkit MCP Servers Explained: Why Most Are Useless (And How to Fix It)

If you want to know about MCP with LLMs, this is the best guide I've seen and one of @vfarcic.bsky.social's best videos.

youtu.be/7baGJ1bC9zE?...

Intro to Lyons and Co creating the first mainframe. The page has been defaced with a Bakewell tart and battenburg cake.

I didn't know the tale of Lyons and Co creating one of the first mainframes! It's worth checking out the book just for this story.

At PARC, Taylor had to find a balance between directing his people and letting them get on with their jobs. Back then there was no management theory that explained this. A theory that explained PARC arrived in 1993 (followed by section on Peter Drucker)

The section on Xerox PARC is great, and Jamie is at his strongest and most insightful when talking about management theory

Just read the best chapters in the book so far, on how photocopiers and tea companies led early computing.

Occasionally dad will lend me a tool and I'll look at it a bit funny and realise it's from the 1960s or even earlier.

I was going to say it's not the same with computers but then I guess the equivalent is using "find" or a similar ancient UNIX tool...

My dad's a joiner so I know enough to know I hate woodworking. I have 0 synths and my guitar playing is bad. I've never made bread though...

Television, the telephone, radio and vinyl records were all analogue technologies.

Interesting point about the first major electronic technologies all being analogue

We're getting to the first computers now. Which apparently calculated how fast a train could go before your head exploded. I don't know the answer, but I did a drawing.

You need it for bluesky DMs as well. I just let it take a photo of my face, which is on my profile anyway.

Claude Code: tokens go brrrr....

This is art.

UPDATE: OpenAI is removing the feature following my reporting, and is working to remove indexed chats. "We think this feature introduced too many opportunities for folks to accidentally share things they didn't intend to," admits OpenAI's CISO.

ChatGPT suggested House of Leaves or Underworld

What book closed the 20th century following Frankenstein and Dracula? Infinite Jest?

Page from visionaries with the text: effects of artificial intelligence will be more transformative than anything that came before it, we should be panicking.

Jamie tells us we should be panicking over AI (some of us are I would argue)

Page from visionaries with observation on timing of Frankenstein and Dracula

Interesting observation (credited to Jeanette Winterson) that the 19th century opened with Frankenstein and closed with Dracula!

Page from visionaries with the text "The art if getting the best out of people is called management. Edison was good at it." highlighted

Page from visionaries on how Edison normalised failure. And a badly drawn Edison light bulb

One of the best insights so far is praising Edison's management and leadership ability as much as his creativeness.

What’s a technology that you think is overhyped? I’m going to give a sideways answer to this, which is that the venture capital business model needs to be understood as requiring hype. You can go back to the Netscape IPO, and that was the proof point that made venture capital the financial lifeblood of the tech industry. Venture capital looks at valuations and growth, not necessarily at profit or revenue. So you don’t actually have to invest in technology that works, or that even makes a profit, you simply have to have a narrative that is compelling enough to float those valuations. So you see this repetitive and exhausting hype cycle as a feature in this industry. A couple of years ago, you would have been asking me about the metaverse, then last year, you would have asked me about Web3 and crypto, and for each of these inflection points there’s an Andreessen Horowitz manifesto. It’s not simply that one piece of technology is overhyped, it’s that hype is a necessary ingredient of the current business ecosystem of the tech industry. We should examine how often the financial incentive for hype is rewarded without any real social returns, without any meaningful progress in technology, without these tools and services and worlds ever actually manifesting. That’s key to understanding the growing chasm between the narrative of techno-optimists and the reality of our tech-encumbered world.

Stand by this: www.politico.com/newsletters/...