ncko retweeted
We are witnessing the fall of a civilization in real time.
ncko retweeted
On Saturday night at the Ellen Melville Centre, philosopher Peter Boghossian @peterboghossian put a sharper diagnosis on the free speech debate than the one we usually hear.
The problem, he argued, isn't only that people are being silenced. It's that people are no longer being honest. Faced with prevailing orthodoxies, most of us learn to say what we think we're supposed to say, rather than what we actually believe. We hedge. We soften. We nod along. We tell pollsters one thing and our close friends another.
That, Boghossian suggested, is the deeper crisis underneath the free speech one. Laws and policies matter, but they're downstream of a culture in which honest disagreement has quietly become socially expensive - and where the cost is paid first by the people closest to the dominant view, who learn to keep their doubts to themselves.
The consequence isn't just personal discomfort. It's that public debate stops doing its job. Bad ideas don't get tested. Good ideas don't get sharpened. Institutions lose the ability to course-correct, because the people inside them are no longer telling each other the truth.
His prescription was unfashionably old-fashioned: be willing to say what you actually think, in plain language, and extend the same permission to others. Free speech as a legal right matters. But free speech as a daily practice - the willingness to be honest in rooms where honesty is awkward - is what keeps a free society alive.
A useful frame for the work ahead.
ncko retweeted
"My first instinct was exactly what such people would delight in happening. To watch the rest of us recoil, then retract, and perhaps even eject. To leave the internet for a while or forever. But I can't do that. We shouldn't do that." world.hey.com/dhh/words-are-…
ncko retweeted
BREAKING: Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all.
They just memorize patterns really well.
Here's what Apple discovered:
(hint: we're not as close to AGI as the hype suggests)
ncko retweeted
being good at competitive programming doesn't mean you are a good swe
like, how tf did we arrive at this conclusion?
ncko retweeted
“If we begin from the position that we’re wrong, and we simply try to be less wrong, we gain more awareness of our blindspots and become less wedded to our beliefs, reducing our resistance to learning.” — @G_S_Bhogal
ncko retweeted
"Do you want to know who you are? Don't ask. Act! Action will delineate and define you."
ncko retweeted
5. Naxalt Fallacy:
Smart people tend to use qualifiers like “generally” and “most”, and dumb people tend to ignore them.
“Most people who are pro-choice are also pro-gun-control.”
“Wrong! I’m not!”
“Men are generally taller than women.”
“False! My wife is 7 feet tall!”
ncko retweeted
4. Package-Deal Ethics:
Being pro-choice and pro-gun-control don't logically follow from each other, yet those who believe one usually also believe the other. This is because most people don’t choose beliefs individually but subscribe to “packages” of beliefs offered by a tribe.
ncko retweeted
In the past, I used gtest for C code. Works ok but cmake and C++ are sort of overkill, and not high level. Next time I unit test C code I'd probably write the unit tests in Python.
Here's a nice article on the idea. What has been your experience?
p403n1x87.github.io/running-…
ncko retweeted
Lots of people in CS are (almost surely) GPT-ing their peer reviews
arxiv.org/abs/2403.07183
ncko retweeted
Navy Pilot blinks morse code warning as his captors instruct him to state he has been treated well. Vietnam, 1966
ncko retweeted
The modern behavior I least understand is wearing a smartwatch
I'm fighting a daily battle to keep my attention from splintering into a million pieces as I'm hunted relentlessly by predatory tech algorithms...
And you want me to tie a Pavlovian training device to my wrist?!
ncko retweeted
C is not just a bad language. It is a demonic instrument, that leaves scars on your soul that not even time can heal.
We had a small bug recently where one extension that works with SQLite stopped working with libSQL. One of our engineers, @iavins, spent a whole day debugging it, and manage to find the commit that broke it.
He looked at the commit puzzled, and told us it wasn't really doing much, just adding some functions and moving stuff around. I just heard the description and uttered: "something, somewhere, changed the size, alignment or field ordering in a struct, and broke everything".
Lo and behold: I was right. As it was night for @iavins in India, I looked at the commit, saw which structs it was touching, and added void pointers to the structs. Just the void pointers, without the rest of the code. And then we found out that one of them, in fact, had to be kept untouched.
v later found out with more debugging that it happened because the module in question cast that struct back from memory, and then added its own stuff to it. So any changes to it are breaking changes.
Mind you: @sarna_dev followed the best practices we learned from our Linux days where everything is done in small, self contained commits. Sounds overkill for these days, but had we committed everything in one big patch, it would have taken days for us to figure out what was wrong.
I code very little these days, now that @tursodatabase is in a more advanced stage. And I haven't coded C in 10 years. But I didn't reason this was the issue. I didn't have to. My body just uttered it. As an instinctive reaction, as an abuse victim to the sight of the abuser.
It is something that just becomes a part of you, and never leaves. It just haunts you forever. It kills your innocence, makes you jaded. Life is never the same.
That's C.
