mnsc

Rigid implies you are trying to change something into something else in a controlled way. Eg a refactoring that does nothing but enable for a future feature. Agents does not change code to refactor anything, they write code to solve the new feature directly. And if you are good at prompting they might overwrite code that exist so that the end result looks like a refactoring. But no clay was reshaped.

Double the waste, double the fun!

"naturally"...

So you are saying that there's a widespread problem with people looking to buy a car for $100k but realizing there's a $6k tax they go, "I can't afford that" and instead buy a unsafe truck?

I lost the link to that short story about humans in the future whose job it is to read and interpret Ai output like it's aliens. Good story! Anyone have the link?

That sounds nice, care to share the breathing exercise?

A recent thing that caused a perspective shift for me was identifying where my drive was coming from. Triggered by the difference between the etymologies of amateur (one who loves) vs connoisseur (one who knows).

Amateur - learns construction on free time ad hoc/by doing, builds a cabin, enjoys living in it

Connoisseur - reads lots of books on construction, might build a cabin to verify competence, enjoys thinking about it

Professional - went to school and works in construction, enjoys the paycheck, doesn't have a cabin

I haven't read the motorcycle zen book yet, but I imagine that is the loving amateur angle, whereas for me I just thrive in knowledge. Translating that to the current AI hellscape, I kinda like that the code crafting bit is a bit less pronounced where I can instead focus on getting to know the system/model/domain of my work. It helps with the dystopian feelings I sometimes get from being in the mine. But if you are a passionate coder, I would imagine that letting someone else build your cabin would be extremely provoking.

Thank you for this post.

My feeling? I ask myself the same question as I do for for everything nowadays. Yeah this might be ai slop, but is it good ai slop? If it's good, like this, I feel good. If it's bad, I feel bad.

Human effort as a proxy for quality... that ship has sailed. And that makes me feel frustrated.

What was the 'don't read the summaries' book?

But they're reading multiple words overlayed on the 30 second video! And they sometimes write "wtf stupd g* n* lol" as a comment! That gotta count!?!

Well I sure hope that the gemini developers don't have access to agentic development tools because they might read this comment, build the multi slide editing feature using 450M tokens and ship it yesterday removing that Claude edge forever.

People complain about all things all the time (rightfully) but I'm not gonna complain about your complaints now, I will only add that I sometimes enjoy humor full of "shock" and profanity. Still! , After decades of professionals and non-professionals diluting it. This site was funny. Ymmv

Lukewarm turkey.

Resistance to change is very different than reluctance of change.

Yeah I read it and didn't get the point. So agile is dead? And we are now doing waterfall due to Ai needing unambiguous specifications? But also agile was nothing new because they knew that waterfall didn't work in the 70s. But now it works? Or are we still doing iterative development but the term agile is banned as unoriginal?

Oh no, the kids are going to invent agile again aren't they?

Nazis have no desire to be part of any democratic system so engaging with them is ultimately an act turning your back on democracy itself. Popper out.

Well I wasn't talking about AR/VR tech. I was talking about a society where using VR like in the the visions the company Meta was drawing up. It hasn't manifested and since the company Meta is now downsizing their VR investments the progress of that AR/VR utopia is not _currently_ getting better, because people don't like it/use it. And I got a kid that was HOOKED on vr and spent so much time with his friends on VR, and now. They hang out on Discord and play CS2 on a very flat 2d screen. Will this change "In The Future(tm)"? I don't care but my point is that a statement like "code quality will be irrelevant" miiiight age poorly if in five years, taking into account model improvements, people still care about code quality and the agentic engineers still use extra effort to make certain the outputted code adhers to "good code" standards. I know I will but if ass code turns out to over time be objectively better than good code, I might change my mind or maybe start a new career in something tangible.

See https://simonwillison.net/guides/agentic-engineering-pattern...

But hindsight is 20/20 as they say. In 2020 people predicted that Facebook Horizon would only go one direction, always improve and become as pervasive as the internet. So when you predict that the design and architecture capabilities of models will continue to improve, thus making code quality irrelevant, you sound very confident. And if in five years you are right, you will brag about it here. If not, well I for one will not track you down and rub it in your face. Peace out.

Me too, why do they do that?

Is this your alt account?

https://news.ycombinator.com/item?id=47222705

Edit: cool article, I have myself speculated that we will get a new language made for/by llms that will be torture writing by hand/ide but easy to read/follow/navigate/check for a human and super easy for Llms to develop and maintain.

I wonder how much the roots of erlang is showing now? Telephone calls had a very specific "natural" profile. High but bounded concurrency (number of persons alive), long process lifetime (1 min - hours), few state changes/messages per process (I know nothing of the actual protocol). I could imagine that the agentic scenario matches this somewhat where other scenarios, eg HFT, would be have a totally different profile making beam a bad choice. But then again, that's just the typical right-tool-for-the-job challenge.

I love that the tool in question is very calm and collected, in contrast with the emotional wreck that is the US regime. I got a very helpful response to this prompt and I will make it continue working on a python script to get my historical chats looking good in Obsidian.

> Ok. So I'm cancelling the subscription to ChatGPT and moving over to Claude because of the news of OpenAI striking a deal with us department of war. (https://www.techradar.com/pro/openai-just-signed-a-huge-deal...) Please line out a good exit strategy where I can keep the information in my chats and projects on my own hard drive.

What would help so much would be to force the platform to act in a way to not try to retain the user at every cost.

Like YouTube. I would love to have parental settings, "no shorts", "no recommendations on video page". Kids could search and click on the creator to see more. Combine that with blocklists and curated allow lists. Boom, YouTube is not a problem. And kids getting bored of clicking around would be a feature.

Conspicuous consumption? Like always?

I'm in the later part of the game and I feel really stupid. Some levels are so small I feel like I can understand all possible strategies but none work. Lovely game overall though, highly recommend!

Is it non-American all the way down?

I think you are reading to much into kids trying to break norms and trying to be "part of a clique". It's not pretentiosness, it's part of finding yourself. They are also actively trying to get you to not read them because you are old and think they "are not serious" so mission accomplished I guess. And time will tell if these kids will invent something you have to respect. (Spoiler alert, we did and they will to)

I read the whole thing without seeing it. But I also fail to see how you marry "pretentious" with not wanting to be taken seriously. So maybe it's me.

But the click bait was conceived and written super fast by a daring driver!