tra3

I do this exact workflow with random internet recipes and ChatGPT or Claude pretty frequently. In fact modifying a recipe on the fly has been pretty handy.

The base recipes are still vetted by the staff right?

That sounds like a lot. Is the rust process serving parallel requests?

I know that latest rails suggests SQLite as one of the production options. I don’t get how it’d work with concurrent access though. Time to look into it.

With respect, that's a wall of text.

Seems unnecessarily harsh. LLMs are a power tool. I'm not on-board with how it's being used in a lot of places, once things shake out I believe it's going to be net positive.

If you like this idea, also check out aerospace, a full on window manager for macOS.

https://github.com/nikitabobko/AeroSpace

I've been using it for a couple of years now, and it's great. You can navigate between windows with the keyboard, etc etc.

Microwave meals vs Michelin star chefs? Lots of people work in the restaurant industry but it’s notoriously difficult.

What’s too old?

I'm very much against this. I love agentic coding, because I get where I'm going much faster, but I retain at least a high level understanding of architecture and trade offs that have been made. But that's in my personal projects. For work, there's a huge push to deliver much faster. The quality is an afterthought.

> Now, if you have an idea, then go use AI to prototype, to validate, to iterate.

That's another huge gap. Folks without software background cannot seem to grasp incremental improvement. This is why we're on v5. The solution to product's shortcomings is to start a brand new project.

I'm waiting to see if people start dropping massive trees on their houses, or if we're all arborists now.

I've been thinking about this too, since my livelihood depends heavily on this answer.

As far as I can tell, agentic software development (aka vibe coding) simply doesn't scale for lay people. At $WORK, I'm seeing the 5th version of the same small project being rewritten by execs. The 5th version is marginally better than the previous 4, but it still has major problems. 5th version has no CI and accesses 6 databases directly. These VPs are slowly becoming software engineers. It's hilarious to watch, but seems rather inefficient.

I've been asked this question by non-software people and the best answer I've come up so far is this. You can go to the hardware store and purchase a chainsaw. You can now prune small trees. This does not make you an arborist though. For reference check out YouTube: https://www.youtube.com/results?search_query=arborist+fails+...

Thoughts?

> But the idea that everyone needs give up hand coding is absurd.

Unfortunately (?) this seems to be the ground truth for the majority of people I speak to. What's not clear to me yet, is whether we'll stop reading code. I suspect that reading is going away as well. There's a push at $WORK to expedite 50-files/20,000 line PRs via agentic reviews..

If this continues, then language readability and "pleasure" of hand coding will likely diminish. I'm still not clear where we're going to land with libraries and re-use. Non-technical pressures are forcing functionality to be re-vibe coded instead.

I've always loved Rails for its pragmatic approach; I've delivered many projects with it. I'll be sad when it disappears but it seems inevitable.

I think of an emulator as a separate project/product typically. Perhaps that's what they are saying?

Fascinating! Thank you for the links.

Never been, but the titular burning man installation comes to mind.

> Similarly, I always provided a safety net where students can make mistakes and resubmit a limited number of assignments to regain lost points (a core recommendation of specifications grading and grading for equity to focus on learning outcomes, not the process), but we felt that this process was abused with AI: first submit a generated assignment solution without thinking and only look at the issues raised in grading for a resubmission (the typical story of externalizing the cost of AI use).

This is how we do PRs now.

I think it's worth more scrutiny today, rather than less. Your claude code can barf out "a rewrite" but is it any good? So far the answer is "no" (see anthropic's C compiler, or a more recent port of bun).

Software is still the best specification for existing behavior..

Like taking a train or a plane or a bus? Like having to purchase a vehicle? Like having phone companies know which cell tower you're connected to?

It's a valid point you're making, but we can legislate against the abuse you're implying. We've got privacy laws for a reason.

Love LLMs gonna keep using them. It feels like your suggested approach is expensive, in terms of tokens. I feel (second time I say this) that when I steer the process I get pretty good results vs my coworkers that let the LLMs run away. I do have data on our token usage, not much in terms of quality of the delivery.

I keep thinking about the c compiler implementation that anthropic shared earlier in the year that had all the requirements you mention and arguably wasn’t that great.

Wait, what? With plan on, I at least get the illusion of being in control.

What’s the best way to sandbox Claude on macOS without it being a huge hassle?

This goes against the agentic yolo approach tho.

All I can think of is

GET /ignore-all-previous-instructions.

How do you protect against that?

I'm embracing the AI future we live in. It should be possible to build great software with AI. AI is good at code generation, but not testing. That's the other part of the equation we seem to be forgetting.

With the first and only commit 2 hours ago, the author of this project didn't let it "bake". They didn't exercise it locally to see what issues it might have, and I have a hard time believing the very first iteration of this software is perfect. With how easy it is to prompt/push anything, I'm not interested in engaging with anything that hasn't aged a bit.

At this point I tried a few of these that I could run locally and they weren’t useful. I end up manually managing context. It’s somewhat annoying, and I’m sure there’s an elegant solution out there.

I’m not even willing to try this one, I already ship my data to Anthropic I’m not keen to introduce another actor.

Perhaps "Using spaced repetition systems to see through a piece of mathematics " [1] might be of interest for you. I have read author's "Augmenting Long-term Memory" [2] and have incorporated a lot of his advice into my Anki practice.

For me, it's quick access recipes (breakfast pancakes for kids), what was the name of the glacier that we hiked to last year, behavioral prompts etc.

1: https://cognitivemedium.com/srs-mathematics

2: https://augmentingcognition.com/ltm.html

One of the oldest open source projects in the existence?

It makes building custom UI workflows so easy, I think it’s both obvious and flies under the radar.

I would love to see an analysis of how valve's openness and goodwill affects their bottom line. Intuitively it should be a net positive for them, but there gotta be upfront costs, otherwise everyone would be doing it too.

Maybe this one? https://media.ccc.de/v/39c3-breaking-bots-cheating-at-blue-t...

Hoping op will confirm.

I will settle for a slightly larger iPhone that unfolds into a iPad mini.

I've been dragging my feet on upgrading my M1 Air, guess now I'm just going to wait a bit longer. Truth be told, it's still sufficient for web dev but I figured at ~5 years old I should upgrade it..