duggan

Hot dog stands don't often masquerade as creative expression.

Whatever they are! Using the literal word "brand" in public messaging is leaking the marketing team's lingo to customers.

For me it's when companies talk about themselves as brands.

Fabulous use of meta-irony, bravo.

Maybe! Too early in the LLM rewrite era to tell whether this requires a LeBron or just someone who takes the local pickup game more seriously.

I do like these sorts of things; decoded a less exciting one from a bottle of wine I found in 2019. Significantly more eye-watering without OCR: https://xcancel.com/duggan/status/1130920846304993282

Interesting. Been to South Africa a bunch but never clocked that they were different from the more common antecedent, the tetrapod[1].

1: https://en.wikipedia.org/wiki/Tetrapod_(structure)

As part of our ongoing engagement with the U.S. government, we previewed our plans and the models’ capabilities ahead of today’s launch. At their request, we are starting with a limited preview for a small group of trusted partners whose participation has been shared with the government, before releasing more broadly.

The clowns in the US administration can barely remain coherent from one sentence to the next.

Having them be the gatekeepers of technological progress in 2026 is fucking lame.

This person should head up writing the next Silicon Valley.

Takes all the fun out of it.

First nobody sane want to give their domain IP to OpenAI/Anthropic. That's why local AI will eventually prevail and flourish because people who actually have some IP will have no problem to buy 10k+ EUR machine to run some pretty good models on it. However if your main job is just doing CRUD stuff, then you are screwed

Replace OpenAI/Anthropic with AWS and this is not too dissimilar to the arguments in 2009 about cloud providers.

It’s not that there's nobody for whom this is true, it’s just that there’s enough of everyone else to build an empire with.

- FileTranscriber, for getting Whisper transcriptions of media files from the context menu on macOS: https://github.com/duggan/filetranscriber

- usagi, for tracking Claude usage in the macOS taskbar. Lots of these around, but I wanted one that wasn't buggy and constantly adding features: https://github.com/duggan/usagi

- RockstarNinja, for sharing Claude plans and sessions. Since it's a bit of a data hog signups are limited, but me and my cofounder use it all the time: https://rockstar.ninja

- TweetEmbed, for browsing an offline Twitter archive and getting self-contained HTML copy/paste-able cards: https://duggan.github.io/tweetembed/

- bewitch, a terminal based metrics collector and data visualization system: https://bewitch.dev/

Each one of these has been a great way to really push on automating build processes, see what Claude can do for automatic documentation (screen captures, etc), and trying to give a distinct visual identity. In my next project, I'm trying to de-LLMify the prose it generates by using my own blog posts and aggressively pruning and integrating into the prompt.

The biggest change has just been unlearning "cost". Stuff I've subconsciously shied away from since intuition built up over a career has given me a sense of how long things take that just isn't true anymore. Still learning!

I used JetBrains IDEs for years. The refactoring tooling was fancy find-replace as far as I recall? Pretty cool in 2015, though.

LLMs are a force multiplier, not magic. They benefit from good tooling.

The bar for "create software" up to this last year or so was "learn software development" or "pay someone else".

Personally, I think millions more people having the ability to create some subset of software is an incredible shift.

$200/month is already out of reach of the majority of the population.

1. You can build small applications with the $20/month sub, much more with the $100/month. Competition and technology improvements will inevitably improve the price to value ratio.

2. Cable sports subscriptions are in a similar price range. Expensive, but not exclusive to “the elites”.

There are some enchanting non-JR regional trains you can get in Kyushu, too.

Got the Shimabara line[1] last summer and it was this tiny cute little single car sailing through farms and along the coast.

1: https://en.wikipedia.org/wiki/Shimabara_Railway_Line

I worked at a startup, they built their API in PHP because it was easy and fast. Now they're successful

You can stop there! Sounds like PHP worked for them. Already doing better than 90% of startups.

We're playing with a fire that catches and spreads so fast, by the time anyone realizes the forest is catching and starting to react, the entire forest is already well on the way to joining in the blaze.

I suspect this has been said in one form or another since the discovery of fire itself.

Very much on the same page as the author, I think AI is a phenomenal accelerant.

If you're going in the right direction, acceleration is very useful. It rewards those who know what they're doing, certainly. What's maybe being left out is that, over a large enough distribution, it's going to accelerate people who are accidentally going in the right direction, too.

There's a baseline value in going fast.

Liz is one of the most genuine and thoughtful people I ever worked with. The software world would be a better place if more people like her were running the show. Best of luck to her.

I did give additional context in the blog post I linked, but yes, to be clear, this is something that will really work best for small projects with reasonably fast build cycles.

If you're already at the point where you're fielding pull requests, lots of long running tests, etc., you'll probably already know you need more than git over ssh.

It does, you're just running a command over ssh, so if you've a particularly long build then something more involved may make more sense.

A directory over SSH can be your git server. If your CI isn't too complex, a post-receive hook looping into Docker can be enough. I wrote up about self hosting git and builds a few weeks ago[1].

There are heavier solutions, but even setting something like this up as a backstop might be useful. If your blog is being hammered by ChatGPT traffic, spare a thought for Github. I can only imagine their traffic has ballooned phenomenally.

1: https://duggan.ie/posts/self-hosting-git-and-builds-without-...

Journalists, anyway. I think I originally heard it from Casey Newton on Hard Fork, but it was a month back so not 100% sure.

But there's loads of people who would be stumped by a for loop, yet can easily work their way through a setup guide, particularly with the hype/promise and an active community.

I've heard it described as the first time many non-programmers have been able to make computers "do things" without it being defined by someone else (app interface, developer, etc). It's a hugely empowering development from that perspective.

The stuff you've listed are the kinds of things smart home enthusiasts do with whatever tools are available to them, and are just a sign of people exploring the possibility space.

I've been tinkering away on one of these myself, https://rockstar.ninja. I expect there are a hundred others out there, going to be interesting to see what the end shape of these tools is.

Motivated novices will just learn differently, and produce different kinds of systems for different audiences with different expectations.

Some will dig into obscurities that LLMs don't or can't touch, others will orchestrate the tools, Gastown-style, into some as-yet-unknown form.

People will vibe themselves into a corner and either start learning or flame out.

I’ve used Claude to write custom firmware in Rust for an ESP32-driven desktop clock.

Turned it into a Stripe revenue dashboard and notifier.

Even bought a couple more, flashed them, and gave to my cofounders, complete with AI written (personally tested, though) setup instructions!

Like any service outage out of their control, people will find other things to do until it’s over.