i don't agree — i am way more optimistic
RE: https://social.vivaldi.net/@lproven/116217029716302392
Can LLMs think is a hard question not because we don’t know how LLMs work. But because we don’t know how “think” works
A.I. Isn't People
https://www.todayintabs.com/p/a-i-isn-t-people
<- We know EXACTLY how LLMs work. It is not that complicated. You can express it in 200 lines.
They absolutely, canonically, by definition, are not aware. They cannot think. This is not controversial or hard.
My turn to pen a blog about my discomfort with AI-generated code. I feel this is a measured take, and addresses something I haven't commonly seen mentioned elsewhere. And it helped me process a bunch. 🙂
"On Making"
Vibe coding works really well when you understand what the machine is doing.
Edit:
It works.
Has the exact feature set and experience I want.
The code isn’t half bad and better than I would’ve written in one evening.
I am the only person using it.
Aside from the ‘learning to program’ argument, explain to me how this is a bad thing.
uname -sr
sysctl hw.versiontime (x11perf \
-dot \
-create \
-destroy \
-map \
-rect500 \
-aa24text \
-copywinwin100 \
-f8text \
-fitext \
-gc \
-getimage500 \
-pcircle100 \
-polytext \
-putimage500 \
-rect500 \
-resize \
-scroll100 \
-tileftext \
-tr10text \
-triangle100 \
-repeat 2 \
-time 1 > /dev/null)
@hi I get complaints about the x11perf arguments not being correct?
@hi likewise, no hw.version on either my FreeBSD or OpenBSD machine
@hi looks like the version on FreeBSD had issues, but on OpenBSD they worked.
Granted, my hardware is ancient, so running it on my ancient Dell Inspiron1420 gives
OpenBSD inspiron1420.attlocal.net 7.8 GENERIC.MP#4 amd64
2m53.26s real 0m06.23s user 0m06.14s system
and I don't even want to try on my VESA netbook (hours maybe? 😆)
@hi the netbook (a Dell Mini10) is OpenBSD (also 7.8) and reports
hw.version=A05
I suspect the presence of that is based on some BIOS value or dmidecode-type output, so it's available on some machines but not others
OpenBSD 7.8
ThinkPad T440s
2m01.70s real 0m06.72s user 0m07.04s system
also curious if recent changes help in any way...
@hi I noticed better X11 overall performance using Radeon card. I think it’s because all stuff is already packed and available in Xenocara. But CPU seem to be less stressed.
If you have an Intel GPU, you should install the "media" package. But this only helps for video rendering, I think. It doesn’t help having the CPU no go on fire when all the ˋcvs up or ˋtar xvf outputs are displayed in xterm.
@hi I meant the one packaged for OpenBSD 🤣
https://codeberg.org/openbsd/ports/src/branch/master/graphics/intel-media-driver
But yes, that’s what I was referring too.
@hi
$ uname -srvm
OpenBSD 7.8 GENERIC.MP#284 amd64
$ sysctl hw.{vendor,product,version}
hw.vendor=Framework
hw.product=Laptop 13 (AMD Ryzen AI 300 Series)
hw.version=A5
$ time (x11perf ...)
2m07.02s real 0m06.21s user 0m06.44s system
Here's a thing that might help that aspect though:
You *do not* need to have an opinion on everything.
The idea that "you should be informed" is a bad one in isolation. Sure, we'd all like infinite knowledge - but no, actually.
Be comfortable with "I have no opinion on that". Does the *topic* actually matter to you? Ok, then go look harder. Until then? No, you shouldn't have an opinion.
Don't let yourself be recruited into some agenda because someone else said you ought to have a viewpoint.
I believe NTPD should be completely disabled by default. It's snithcing about yourOS. Sending information to linode.domain. We need more privacy. https://marc.info/?l=openbsd-bugs&m=177296357231841&w=2
why some of the files listed twice? o_O
#openbsd
There’s one very simple thing I learned from Chomsky and that will change your worldview forever. It is really simple.
Each time you read something in a media, simply add the following sentence before everything: "Someone want me to believe that…"
It completely modified my perception. It calmed my anger against lies. It allowed me to read something with no judgement other than "Someone want me to believe that".
Next, you will start questioning who is that "someone". But one step at a time…
I moved all my repos to #GotHub:
And you can do the same! Go to https://gothub.org/ and check it out!
Maybe you are interesting on this one too:
for whoever needs to hear this: you're not alone. i'm not vibecoding any of the software i write. i'm writing it by hand, but i've leveled up my emacs with eglot/lsp. i'm modernizing my stacks and use languages with excellent compilers. i think about how to do more with less. i'm trying to combine the best human-written libraries and modules and assemble them with minimal boilerplate. i enjoy reading your manuals and references. i believe in robust, secure, human-written software.
I really appreciate all the Mastodon mentions and camaraderie for this essay. ❤️ I suppose I really should toot it myself:
Friendly reminder of some really good resources to lean email-based git workflow when your centralized forge goes down:
Authoring & Sending Patches
https://git-send-email.io
Receiving & Applying Patches
https://git-am.io
Gotta give it to Claude Code developers: they are exceptionally brave. I would’ve been straight up embarrassed to show stats like this.
Reminder: Claude Code is a thin CLI client that reads input from your terminal, sends it to Antrhopic servers and prints back the response.
it's like asking contributors to edit files in your favorite text editor only... and your text editor is microsoft word
'' and sometimes double quotes "" (i almost always use emdash)We're happy to announce a long-term partnership with Motorola. We're collaborating on future devices meeting our privacy and security standards with official GrapheneOS support.
https://motorolanews.com/motorola-three-new-b2b-solutions-at-mwc-2026/
The LLM topic has been all over my mastodon feed for months. I find the consequences of LLM adoption depressing overall, with all the damage resulting in several segments of our societies worldwide.
Until now, I have been ignoring LLMs, but there is increasing use of LLMs among customers of my company, which means I can no longer ignore this topic entirely.
I observe use of LLMs mostly by people who don't write programs regularly, who are using these tools to fill gaps in their own skills or available time, with variable success.
The only work item related to LLMs I have accepted so far is reviewing LLM-generated security bug reports, where someone else is running various AI tools to scan open source projects, sends us reports, and with respect for our time (unlike some other people who just spam open source projects with such reports) pays me and another open source developer to take a look at them.
Most of these reports are garbage and get discarded. About 1 or 2 in 25 reports are on to something. We write required fixes the good old fashioned way.
I have been reviewing reports from code scanners for more than a decade every now and then. The only thing which is new to me here is the entanglement of the code-scanning tool with all the harmful side-effects and consequences of its existence.
I haven't yet received significantly higher quality reports than what I have seen before LLMs. A big problem is that the severity of the bugs reported is often blown out of proportion, which can cause wrong judgement or even panic when non-experts are evaluating such reports without a sufficiently critical lens.
Reluctantly setting aside the larger issues surrounding LLMs, code-scanning is as far as I will accept going along with this, but no further.
My company is now borrowing the EU's "Certified Organic" logo to deter potential clients who would require use of LLMs. I hope this gets the point across, without having to explicitly mention LLMs or "AI", cause I am very much sick of seeing them mentioned everywhere.
I've moved some of my currently active repositories over to Game of Trees Hub! 🌳
GoTHub is a transparently funded Git/Got repository hosting service - lightweight, BSD-licensed, and a great minimalist alternative to the big tech.
Check it out: https://rsadowski.gothub.org/
i wrote a little time log parser in posix shell. it collects from all the files in my todo directory, selects all the lines formated like this...
# start-time stop-time project: task description...and outputs logged time grouped by projects
- 20260301-085700 20260301-095640 code: add log.sh
code 05:10:44#shell #journal
home 00:22:00
kids 00:42:40
meta 00:40:22
best answer: man man my man 🙂
CfP is coming up for EuroBSDCon 2026 in September!
https://2026.eurobsdcon.org/
17 new VMs were added and 68 VMs were renewed.
We donated €1190 to the #OpenBSD Foundation, €63385 since we started.
Thank you, our users, and OpenBSD developers for an awesome OS!
Stay safe & healthy!
#RUNBSD in 2026