hansmayer

> you can see that it's doing exactly this inspection when the skill runs

I mean how do you know what does it exactly do? Because of the text it outputs?

Ed Zitron's latest piece has a great take on this - basically yes, they thing they've unlocked a great secret and they think they are very smart, when instead they are actually doing the work for LLM, while giving LLM the credit for the outputs of their work.

:) I keep saying it - the AI will cost us all dearly, but not in the ways the AI boosters are saying it will....

> Spec-driven approach is fun

...If you never ever look at the code that's generated, it probably is.

Great points, but I imagine it's a bit too heavy on the rigorousness requirement for the LLM crowd. The folks are high on this stuff and I am beginning to notice it's like trying to get a heavy pothead or crackhead of off their stuff. Don't you see it - if you just wave your hands a lot, and tell the LLM to be serious about it, the scores will just appear :) It's true in their own frame of reference.

Oh dear, I thought you were merely sarcastic in your first comment. But you seem to have been fully converted to the LLM-religion, and actually believe they actually "think" or "know" anything?

> They aren't arbitrary, as I said earlier I got the LLM to de a detailed analysis first, then summarise

I think you still owe us an explanation as to how the score is constructed...

You gotta love the randomly assigned score, like if LLM is actually able to measure anything. But then again, we now call a blob of text a "skill", so I guess it matches the overall bullshit pattern.

> So Kagi gave you ads, sponsored links, AI generated answers etc. as top results? =)

No, I did not register for kaggi.search if that's what you are implying.

This was about ~12 months ago, so no, the AI-generated summaries were not a thing you would expect to see at that time, outside of ocassional A/B experiments. Now maybe my expectation is different to yours. I'd expect I dont have to do much tweaking or downranking at all. My pre-2019 google experience had been roughly "i type in 3-4 words" and one of the first three links is exactly what I searched for. Kagi did not deliver that, much as I would love it to. But with Kagi relying mainly on google search index, it seems to me there is only so much they can do anyway, apart from users ranking and downranking stuff...which I am not keen to do...

This is something I've been curious about for a long time now. I would happily pay their top tier, but testing their free tier did not seem to produce much different results compared to using google ? Or was I not using it right? Just typing in same stuff and literally getting the same list of results. If you want to share more, I'd be happy to know.

No kids, don´t put yourself through this suffering. If you have to invest so much deliberate effort to sort of make it work - while you still handle the most tenuous and boring parts yourself, then what is the point? Lets keep the LLM vendors to their word - they promised intelligent machines that would just work so well to the point of causing mass unemployment. Why on earth do we have to work around the LLMs to make them work? What is the point? Where is my nation of datacenter PhDs or a PocketPhd, depending on whose CEOs misleading statement one quotes?

Oh didn't you know - the legend has it, he was also bad at maths in elementary school, so that kid of your friends, failing the math this year, may be a genius too :)

> I don't think there is that much value in memorizing rarely used, easily looked up information.

easily looked up - we don't have that any more since Google decided to entshittify search. What you now have is not looked up information, as that would look the same each time you "looked it up" - instead of a quick 3-4 word search pattern you now write an elaborate verbose "query", and get a chewed up re-interpretation by the shitty LLMs. And then since sometimes its not quite what you asked for, you have to ask it again or redirect it and just like that, you've wasted 5 mins of your time arguing with a goddamn neural network!

> I was never a big framework user,

So you only ever wrote code in academic setting? Not being sarcastic, but no realistic software development in commercial setting will happen without frameworks.

> That's not how I understood other's experience to be, they're describing something that won't let them even write toy calculators. Selected quotes:

Well that is literally not what they are describing. The man said "boilerplate code". Again, unless most of the code you wrote was code used in purely academic setting, whether research or teaching, we all understand boilerplate code to mean something like "code to open a file descriptor / iterate through database rows / poll web server for results" etc. So the typical stuff you would implement while relying on abstractions and concepts someone else already defined for you.

> What are they worried about then?

If I had to guess - they are worried about the growing LLM-backlash working against their hoped-for-industry-wide-adoption and their infinite money at some point running out, because at the end of the day, NVIDIA lending OpenAI 100B to invest in MS Azure and Azure using those 100B to purchase NVIDIA chips...is starting to look a lot like circular financing.

> sounds like they're worried about "using these and those frameworks" but that's far from "real world work" in my experience

Again, if your area of work is academic or teaching, then this probably matches your own real world experience. The problem for the LLM-crowd is that there is a lot more people here whose "real world experience" are not toy calculators and isolated algorithm implementation, but actually yes, using "this and that framework", as otherwise the implementation would take un-realistically long. I may know how TLS works, but would I implement it's handshake routine on my own? Only if I did it as a personal excercise, not in a commercial scenario. Same for UI, same for DB ORMs, etc. You name it.

Sad, but true.

Mate, we don't trust it to write an email or the code it generates. Why should we trust it to fly a plane?

Well - that's the thing - toy calculators are easy on the "muscle memory". The operations and data types will be mostly similar in syntax across all these languages. If however you wanted to do something more akin to a real world example, using these and those frameworks... it would probably look different. I wasn't disputing the knowledge of multiple languages btw - some of us had experiences in languages from the times way earlier than C# and Java. The point is for real world work you wont quite do toy calculators and the people pushing for "AI writing all the code" are not worried about you retaining the "muscle memory" to write addition and substraction functions....

The minor benefits of vibecoding unusable prototypes or lazy cretins "writing" blogs with AI can't quite compare to the benefits of cars and factories, don't you think?

> random JavaScript, C#, Java, Rust and Clojure "manually"

Right, sounds very credible to me. What did you write, an addition function in each of those?

For the love of all that's holy - folks please stop using AI to publish smart sounding texts. While you may think you are "polishing" your text, you are just disrespecting your readers. Write in your own words.

I'm not going to play your editor here, but, yes I'd prefer an imperfect human-written essay to an AI slop word salad any day. The reason we read essays and blogs is to get a different point of view or even understand something about an author. It's the human point of view I am interested in, not statistically optimised "content" that looks polished on the outside, but is empty on the inside.

> That said, there's a reason.. I'm trying to communicate an idea

Then communicate it - in your own words. Being made to read through automatically generated text which seeks a statistical average makes me livid - it tells of a certain disrespect towards the reader. Slapping AI-slop onto your draft and calling it your work is dishonest.

Here is an honest critique for you: it is too dense, too convoluted, unreadable and lacking that coherent, logical structure that even the laziest human-written article inherently posseses. I was going to read it, but after a few parapgraphs, it just gives off that familiar, yet weird AI-slop vibe. Do yourself a favour and write something on your own.

>and have a lot of additional personal interaction along the way.

Right, well this explanation sounds even weirder, whatever the hell it is supposed to mean. Sick of half-intelligent crap that is pushed out by quarter-intelligent humans "aided" by their slop generating "tools". You guys will hit the last nail in the coffin of the formerly readable world wide web.

I'd be happy if we got to a point where AI-generated content like this was banned. Then we can move onto more pressing issues.

"Gitbutler", really rolls off the tongue, doesn´t it :) Oh and the irony of raising $17M to "replace" a tool which kinda...does not need replacing at all? How about replacing some of the entshittified services, like Google Workspace? Now that would be worth the $17M raised.

Well, colour me surprised :)

"Make this application without bugs" :)

Well, there is a ton of normal CEOs of "normal" companies doing a great job, due diligence and all. Its just these weird CEOs of weird companies with weird business models, who burn hundreds of billions of dollars to produce effectively peanuts, that make a lot of decent CEOs look bad.

Exactly my point here...