This author, like many others on this site, seem to imply that AI generates "good" code, but it absolutely does not---unless he's running some million dollar subscription model I'm unaware of. I've tested every AI using simple Javascript programs and they all produce erroneous spaghetti slop. I did discover that Claude produces sufficiently decent Haskell code. The point is that the iterative process requires you know the language because you're going to need to amend the code. Therefore vibe in the language you know. Anyone that suggests that AI can produce a solid application on its own is a fraud.
The Erlang space vs the Elixir space (can't speak for agent based code generation here) would seem to give credence to this theory.
When I would explore Elixir forums with much larger communities there'd be myriad base level questions with code blocks written as if Elixir and Ruby were interchangable cause the syntax looks similar and thus missing out on many of the benefits of OTP.
But when you'd go to the Erlang community to ask a question, half the time the author of the book or library was one of the like... 20 people online at any given moment, and they'd respond directly. The quality of the discussions was of course much deeper and substantial much more consistently.
I have not tried to generate Elixir vs Erlang code but maybe it'd be a neat experiment to see if the quality seems better with Erlang
You, like many others, seem to imply that humans write "good" code, but they absolutely do not--unless they are running some million dollar team with checks and cross checks and years of evolving the code over failures. I've tested every junior developer using simple Javascript leetcode quizes and they all produce erroneous spaghetti slop.
The difference is, we forgive humans for needing iteration. We expect them to get it wrong first, improve with feedback, and learn through debugging. But when AI writes imperfect code, you declare the entire approach fraudulent?
We shouldn't care about flawless one-shot generations. The value is in collapsing the time between idea and execution. If a model can give you a working draft in 3 seconds - even if it's 80% right - that's already a 10x shift in how we build software.
Don't confuse the present with the limit. Eventually, in not that many years, you'll vibe in English, and your AI co-dev will do the rest.
"This extension uses a fully embedded Neovim instance, no more half-complete Vim emulation!"
This article seems to compare an out-of-the-box vscode with a tailor made vim.config. I guarantee if he spent as much time configuring vscode as he did configuring vim he could establish equivalent environments. That vscode-neovim plugin alone (not vscode-vim) is likely sufficient.
Yes. I don't see the point, though. If I can run full neovim inside VSCode what does VSCode add?
> I guarantee if he spent as much time configuring vscode as he did configuring vim he could establish equivalent environments.
Yes, almost. For example you can make VSCode run terminals in tabs like the editor windows. VSCode supports a lot of customizing. But it still runs on Electron with a rather heavy node process on the other end -- a lot heavier than vim or neovim.
I used VSCode for over six months but ended up going back to vim. Nothing specifically wrong with VSCode, I recommend it to people who don't know how to use vim/neovim, or don't want to use those tools. But for those of us who know how to use vim/neovim VSCode feels slow and bloated with features I don't want. Personally I prefer not to use Microsoft products when I can help it, but now with VSCode (and GitHub) increasingly pushy about AI that I don't want I can do without VSCode.
Very likely the same way that targeted advertising is carried out, which most certainly does occur. Also, 4chan has a history with law enforcement, including the FBI, and the military. Moreover multiple studies have been carried out by academia including by the Israeli military in regard to demographics. The point is that an entity like Palantir could be applied to reduce hate speech and violence. Rather than serve an ad it uses AI to generate a post that is humorous yet very curated in an intrusive insidious fashion.
If I ask Grok about anything that occurred this morning, he immediately starts reading and researching in real time. "Summarize what Leavitt said this morning." "Tell me what's new in python 3.14." Etc.. What do you mean by "cutoff", it seems unlikely that Claude is that limited.
My blog is private, offline, and only kept on my personal local computer therefore the only way for "people" to read it is according to an illegal criminal way. Moreover, in the US it would be a felony. I'm currently using Hugo.
An obvious maxim. However, the full statement is "Privacy empowers YOU, but surveillance weakens YOU." Thus, to the enemy of the people, surveillance is a great weapon. Enemies of the people are typically an elite ruling class. To me, surveillance is generally applied to oppress opportunity, e.g. contemporary cancel culture, bad jacketing, or blacklisting.
It is implied, or goes without saying, that baby loves baby too. A love song wouldn't explicitly state this because it would appear vain and come off as a slight. Thus, the actual consequent is "((x == singer || x == baby) && (singer /= baby))"
But, as also seen with DOGE, they’re careless and refuse to do even cursory research on anything beyond the first-order effects of their actions which, to me at least, seems like a terrible set of traits for being in charge of weaponry.
Musk behaved the EXACT same way when he took over Twitter and wanted to move out of one of the 3 main data centers they used. Actual experts told him it would take a year so he instead impulsively hires basic movers to rip the servers out and was surprised at how many things this broke. He said "I didn't realize how many places that data center was hardcoded". Except people absolutely told him that.
I don’t think he’s particularly frugal. How much money did he blow buying his favorite social network again?
DOGE (and Twitter) are examples of a modern American business management affliction: if you can’t think of anything to create, the next best thing is to start tearing other things down in the name of “efficiency”. Reap the short term benefits and then get out before it collapses.
Only gullible people actually think that. Musk is using DOGE to settle scores and shut down agencies investigating him or regulating his companies or ones he doesn't like.
Everything Musk has done in the past few years has proven he is not to be believed or trusted in any way.
He is a destructive moron. He is the gremlin on the wing of the plane. Between Tesla, Twitter, and now the Federal Government he might have fired more people than anyone else in history.
This is the scene I keep coming back to! Can you imagine having actually gotten into a space ship bound for Mars with this guy [0] ? He'd be whining that the trip was taking too long, complaining that the ship must be too heavy, and trying to drill speed holes in the hull.
[0] (as we envisioned back when we thought this guy might actually put his own skin in the game like that, before his tragic spiral into social media and ketamine addictions)
reply