abalashov

8 ms·

151.
by abalashov 5mo ago
... no. As someone who has had both, I can tell you there's _quite_ a difference.
152.
by abalashov 5mo ago
I take that as a given, too, especially considering the diabolical architecture of Miller Thought.
153.
by abalashov 5mo ago
This is insane. I cannot fathom how I, nor educated and talented people I know, could have possibly stayed in the US back in the day if this requirement had been in place then. Applying for a greencard while working on an H, J or O-class vi
154.
by abalashov 5mo ago
> Once they hit a wall, that is where you find out whether they are motivated or not Yep. That has to happen first.
155.
by abalashov 5mo ago
Because it's "easy"?
156.
by abalashov 5mo ago
Yeah, I know. Was just throwing it out to see what folks would say based on their idea of useful.
157.
by abalashov 5mo ago
They kept telling us Soviet central planning was inefficient and ineffective, while the lean rationality of the free market something something...
158.
by abalashov 5mo ago
How much would one have to invest in hardware to feasibly run a usefully large Kimi K2.6?
159.
by abalashov 5mo ago
Why do you... seem so sure that this is a ridiculous statement?
160.
by abalashov 5mo ago
I am impressed that Graeber made his way into this conversation (and the author's article), and really appreciate your take. I think about "Bullshit Jobs" every time I hear someone tell me how lean, efficient and rational the
161.
by abalashov 5mo ago
I self-host because I'm not training some model for free with our and our customers' proprietary code.
162.
by abalashov 5mo ago
Go is less a language than a philosophy. It was an angry reaction to 10,000 ways to do things, and overly clever (ahem, expressive) syntactic sugar. It is quite boring to write, but very easy to read. Not a Go fanatic. I use Go and various
163.
by abalashov 5mo ago
And they have a suspiciously well-appointed MCP server.
164.
by abalashov 5mo ago
I have an M4 Max MBP with w/128 GB of RAM, and have been very into local models. Qwen3.6-35B-A3B has been very kind to me for almost any purpose. No, it's no Opus 4.7, but it's shockingly good, and I don't have to share
165.
by abalashov 5mo ago
You'd be surprised how far you can get with self-hosted Gitea.
166.
by abalashov 5mo ago
This is an unexpectedly apt comparison, and I appreciated it.
167.
by abalashov 5mo ago
That sort of syntactic sugar goes against the Go philosophy. Don't get me wrong, I share your frustration, but I also see the value of consistency in their philosophy.
168.
by abalashov 5mo ago
I laughed hard!
169.
by abalashov 5mo ago
That may indeed look like quite the speed-up. But the accumulated errors and entropy in such an enterprise will eventually cause a cave-in, at which point the productivity metrics don't look so good.
170.
by abalashov 5mo ago
This is exactly my conclusion, to the letter.
171.
by abalashov 5mo ago
No, humans don't produce 100% reasonable code, but the nature of human mistakes, foibles and unreasonableness is very different from the kind slop farming yields.
172.
by abalashov 5mo ago
It's nice to know I'm not alone in this. I have definitely used slop as inspiration by negative example.
173.
by abalashov 5mo ago
I went back to writing code by hand quite some time ago and cannot say there has been any loss of velocity or productivity for it. I really do think this whole thing is a wash.
174.
by abalashov 5mo ago
I work by myself and feel no joy in using AI.
175.
by abalashov 6mo ago
Great article! I use LLMs 99.9% in the chat box, to bounce ideas around, ask research and reference questions, and so forth. However, I have sworn off 'agents' entirely for serious code, for reasons your article captures almost to
176.
by abalashov 6mo ago
There is an inescapable connection between the two. Knowing that stuff is not _equivalent_ to knowing good architecture, but it's hard to imagine what knowing good architecture would look like without being able to "think in code&
177.
by abalashov 6mo ago
One cannot build something robust just by thinking about it _a priori_, and while this was somewhat at the periphery of the author's argument, it is important.
178.
by abalashov 6mo ago
The problem there is the _large_ language model part, the density and the reinforcement of the weights. There's far less good code in the world. ;) These things emit code as well as I do, such as they do, only because they've inha
179.
by abalashov 6mo ago
> A generated piece of code is linguistically plausible but often unknowingly imitates common idioms so incoherently that the actual bug may be accidentally disguised in a way no sane human (even a bad programmer) could have come up with
180.
by abalashov 6mo ago
You stole what I was about to write out from under me. This is exactly it.

More ›