That's a big leap of faith and... kinda contradicts the article as I understood it.

My experience is entirely opposite (and matches my understanding of the article): vibing from the start makes you take orders of magnitude more time to perfect. AI is a multiplier as an assistant, but a divisor as an engineer.

Yes, I was thinking about context buffers, which I assume are not small in large models. That has to be loaded into VRAM, right?

If I keep sending large context buffers, will that hog the batches?

How large is a full context window in MiB and how long does it take to load the buffer? I.e. how many seconds should I expect my worst case wait time to take until I get my first token?

I assume you gathered a lot of thoughts over these 15 years.

Should I invest in making the switch?

Some highlights that made me think:

> It's easy to say you're faster if you don't actually support everything or maybe even made a mistake.

> I don't see any tests so I wouldn't use this.

---

> the repo has 5 commits and the first one is from 3 hours ago. "I've been working on" is probably more accurately "this morning I asked an ai to write this for me".

---

> The single-threaded design of redis was specifically so that operations are ordered sequentially, so that the WAL-like log would be replayable and you'd get the exact same state as when shutting down the server.

> Did you take any measures to ensure a sequential order of executed commands?

How can you "check" that which you don't "understand"?

> I'm not sure that's how professionals use LLMs right now

I'm a professional and I can tell you how I use LLMs: I write code with their assistance, they don't write code for me.

The few times I let Claude or Copilot loose, the results were heartbreaking and I spent more time reviewing (and then discarding) the code than what it took me to later write it from scratch.

It takes more, not less, time to thoroughly review code you didn't write.

Well, that explains the sloppy results Microsoft is delivering lately!

And this won't change a thing: it doesn't matter if they make a Linux-native frontend to the horrible GOG Galaxy. I just want my games to launch as seamlessly as they do from Valve's UI, not yet another launcher that I have to launch on top of Valve's system UI. I am already doing that with Heroic Games Launcher, which is far better than whatever they will concoct in-house and supports many other stores.

Why is this such a common occurrence here? Does this fallacy have a name?

EDIT: seems to be https://en.wikipedia.org/wiki/Nirvana_fallacy

[0] https://steamcdn-a.akamaihd.net/apps/valve/2007/SIGGRAPH2007...

In Spain's case Telefonica (largest telecom, used to be state owned) is private but has a large State participation and the government literally appointed the latest CEO.

Guess who sells the largest football games as part of their expensive TV package?

Guess who asked a judge to order the other telecoms to also block Cloudflare IPs?

Thus comments are mostly neutral, objective facts that add upon the original comment, or negative comments of disagreement.

Yeah I'm the one in a bubble because I think players that play competitive games expect competitive integrity, regardless of their skill level.